Compare commits
1
Commits
| Author | SHA1 | Date | |
|---|---|---|---|
 luni
|
af9811e731 |
@@ -2,7 +2,7 @@ from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['gpu'] = '4'
|
||||
config['name'] = 'body_torch_cape_m'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'body']
|
||||
@@ -11,9 +11,9 @@ config['script'] = 'scripts/ae_body_train.py'
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/128/cape_processed/train/male'
|
||||
config['valid_dir'] = '/data-local/128/cape_processed/test/male'
|
||||
config['test_dir'] = '/data-local/128/cape_processed/test/male'
|
||||
config['train_dir'] = '/data-local/127/cape_processed/train/male'
|
||||
config['valid_dir'] = '/data-local/127/cape_processed/test/male'
|
||||
config['test_dir'] = '/data-local/127/cape_processed/test/male'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
|
||||
@@ -1,73 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'norm_torch_2'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'norm']
|
||||
config['script'] = 'scripts/ae_norm_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/train'
|
||||
config['valid_dir'] = '/data-local/127/valid'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_mask',
|
||||
'uv_normals',
|
||||
'mesh_mask',
|
||||
'mesh_normals',
|
||||
}
|
||||
@@ -3,25 +3,25 @@ from utils.convnext_utils import CNBlockConfig
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'abl_torch_norm'
|
||||
config['name'] = 'cape_00032_SL'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['iccv', 'abl_norm', 'torch']
|
||||
config['tags'] = ['iccv', 'cape_00032_SL', 'torch']
|
||||
config['script'] = 'scripts/cvae.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['remove_global_orientation'] = True
|
||||
config['use_normals'] = True
|
||||
config['use_curvature'] = True
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_offsets'] = True
|
||||
|
||||
config['use_norm_loss'] = True
|
||||
|
||||
ablation_count = [config['use_skinning'], config['remove_global_orientation'], config['use_normals'], config['use_extra_losses'], config['use_offsets']].count(False)
|
||||
ablation_count = [config['use_skinning'], config['remove_global_orientation'], config['use_curvature'], config['use_extra_losses'], config['use_offsets']].count(False)
|
||||
assert ablation_count == 1 or ablation_count == 0
|
||||
|
||||
config['avg_garment_path'] = 'custom_data/avg/train_no_shape_avg_uv.npy' #custom_data/train_no_shape_avg_uv.npy # train_no_shape_avg_uv train_no_shape_one_hot_avg_uv
|
||||
config['avg_garment_path'] = 'custom_data/avg/train_cape_f_03223_SL_2_no_shape_avg_uv.npy' #custom_data/train_no_shape_avg_uv.npy # train_no_shape_avg_uv train_no_shape_one_hot_avg_uv
|
||||
if config['use_skinning']:
|
||||
assert (config['use_offsets'] and config['avg_garment_path'] != '') or (not config['use_offsets'] and config['avg_garment_path'] == '')
|
||||
|
||||
@@ -30,19 +30,20 @@ config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['dataset_in_memory'] = True
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/train'
|
||||
config['valid_dir'] = '/data-local/127/valid'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['train_dir'] = '/data-local/127/cape_processed/train/male'
|
||||
config['valid_dir'] = '/data-local/127/cape_processed/train/male'
|
||||
config['test_dir'] = '/data-local/127/cape_processed/train/male'
|
||||
config['train_frames'] = 'frames_00032_SL_train_2.txt'
|
||||
config['valid_frames'] = 'frames_00032_SL_test_2.txt'
|
||||
config['test_frames'] = 'frames_00032_SL_test_2.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccv/static_torch_orig_2'
|
||||
config['body_model'] = 'checkpoints/iccv/body_torch_1'
|
||||
config['normals_model'] = 'checkpoints/iccv/torch/norm_torch_1'
|
||||
config['static_model'] = 'checkpoints/iccv/torch/static_torch_cape_m'
|
||||
config['body_model'] = 'checkpoints/iccv/torch/body_torch_cape_m'
|
||||
config['curvature_model'] = 'checkpoints/iccv/torch/curv_torch_cape_m'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
@@ -61,7 +62,7 @@ config['signature']: set = { 'id',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_curvature',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
@@ -81,7 +82,7 @@ config['bottleneck_scale'] = 2
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
config['cvae']['input_features'] = 3+3+3+7 if config['use_curvature'] else 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
@@ -107,33 +108,33 @@ config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = True
|
||||
config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
####################### cond_ae_curv ############################
|
||||
config['cond_ae_curv'] = dict()
|
||||
config['cond_ae_curv']['device'] = config['device']
|
||||
config['cond_ae_curv']['is_label'] = True
|
||||
config['cond_ae_curv']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_curv']['kernel_size'] = 7
|
||||
config['cond_ae_curv']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
CNBlockConfig(128, config['cond_ae_curv']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_curv']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(config['cond_ae_curv']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
config['cond_ae_curv']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_curv']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_curv']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
@@ -2,32 +2,48 @@ from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'cvae_norm_1'
|
||||
config['gpu'] = '4'
|
||||
config['name'] = 'cape_03223_SL'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['iccv', 'cvae_norm', 'torch']
|
||||
config['script'] = 'scripts/cvae_norm.py'
|
||||
config['tags'] = ['iccv', 'cape_03223_SL', 'torch']
|
||||
config['script'] = 'scripts/cvae.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['remove_global_orientation'] = True
|
||||
config['use_curvature'] = True
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_offsets'] = True
|
||||
|
||||
config['use_norm_loss'] = True
|
||||
|
||||
ablation_count = [config['use_skinning'], config['remove_global_orientation'], config['use_curvature'], config['use_extra_losses'], config['use_offsets']].count(False)
|
||||
assert ablation_count == 1 or ablation_count == 0
|
||||
|
||||
config['avg_garment_path'] = 'custom_data/avg/train_cape_f_03223_SL_2_no_shape_avg_uv.npy' #custom_data/train_no_shape_avg_uv.npy # train_no_shape_avg_uv train_no_shape_one_hot_avg_uv
|
||||
if config['use_skinning']:
|
||||
assert (config['use_offsets'] and config['avg_garment_path'] != '') or (not config['use_offsets'] and config['avg_garment_path'] == '')
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['dataset_in_memory'] = True
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/train'
|
||||
config['valid_dir'] = '/data-local/127/valid'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['train_dir'] = '/data-local/127/cape_processed/train/female'
|
||||
config['valid_dir'] = '/data-local/127/cape_processed/train/female'
|
||||
config['test_dir'] = '/data-local/127/cape_processed/train/female'
|
||||
config['train_frames'] = 'frames_03223_SL_train_2.txt'
|
||||
config['valid_frames'] = 'frames_03223_SL_test_2.txt'
|
||||
config['test_frames'] = 'frames_03223_SL_test_2.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccv/static_torch_orig_2'
|
||||
config['body_model'] = 'checkpoints/iccv/body_torch_1'
|
||||
config['static_model'] = 'checkpoints/iccv/torch/static_torch_cape_f'
|
||||
config['body_model'] = 'checkpoints/iccv/torch/body_torch_cape_f'
|
||||
config['curvature_model'] = 'checkpoints/iccv/torch/curv_torch_cape_f'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
@@ -44,14 +60,19 @@ config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'uv_normals',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_curvature',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
'mesh_normals',
|
||||
}
|
||||
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['upscale_enhancer'] = ''
|
||||
config['freeze_encoders'] = True
|
||||
@@ -61,7 +82,7 @@ config['bottleneck_scale'] = 2
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3
|
||||
config['cvae']['input_features'] = 3+3+3+7 if config['use_curvature'] else 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
@@ -75,7 +96,7 @@ block_setting = [
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(176, 256, 3),
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'] * 2, 256, 3),
|
||||
CNBlockConfig(256, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
@@ -2,51 +2,48 @@ from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'abl_iccvw_all_5_10p'
|
||||
config['gpu'] = '4'
|
||||
config['name'] = 'cape_03223_SS'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['iccv', 'abl_all', 'torch']
|
||||
config['tags'] = ['iccv', 'cape_03223_SS', 'torch']
|
||||
config['script'] = 'scripts/cvae.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_normals'] = True
|
||||
config['remove_global_orientation'] = True
|
||||
config['use_curvature'] = True
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_offsets'] = True
|
||||
|
||||
config['use_norm_loss'] = True
|
||||
|
||||
ablation_count = [config['use_skinning'], config['use_normals'], config['use_extra_losses']].count(False)
|
||||
ablation_count = [config['use_skinning'], config['remove_global_orientation'], config['use_curvature'], config['use_extra_losses'], config['use_offsets']].count(False)
|
||||
assert ablation_count == 1 or ablation_count == 0
|
||||
|
||||
config['avg_garment_path'] = 'custom_data/avg/train_no_shape_avg_uv.npy' #custom_data/train_no_shape_avg_uv.npy # train_no_shape_avg_uv train_no_shape_one_hot_avg_uv
|
||||
config['avg_garment_path'] = 'custom_data/avg/train_cape_f_03223_SL_2_no_shape_avg_uv.npy' #custom_data/train_no_shape_avg_uv.npy # train_no_shape_avg_uv train_no_shape_one_hot_avg_uv
|
||||
if config['use_skinning']:
|
||||
assert (config['avg_garment_path'] != '')
|
||||
assert (config['use_offsets'] and config['avg_garment_path'] != '') or (not config['use_offsets'] and config['avg_garment_path'] == '')
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['dataset_in_memory'] = True
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
config['train_dir'] = '/data-local/127/cape_processed/train/female'
|
||||
config['valid_dir'] = '/data-local/127/cape_processed/train/female'
|
||||
config['test_dir'] = '/data-local/127/cape_processed/train/female'
|
||||
config['train_frames'] = 'frames_03223_SS_train_2.txt'
|
||||
config['valid_frames'] = 'frames_03223_SS_test_2.txt'
|
||||
config['test_frames'] = 'frames_03223_SS_test_2.txt'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt_10p.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train_10p.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4_10p'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_10p'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_10p'
|
||||
config['static_model'] = 'checkpoints/iccv/torch/static_torch_cape_f'
|
||||
config['body_model'] = 'checkpoints/iccv/torch/body_torch_cape_f'
|
||||
config['curvature_model'] = 'checkpoints/iccv/torch/curv_torch_cape_f'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
@@ -58,7 +55,6 @@ config['latent_dim'] = 128
|
||||
config['lr'] = 0.0001# 0.0001
|
||||
###############
|
||||
config['starting_balancing_coeff'] = 0.0001
|
||||
config['balancing_activation_limit'] = 0.01
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
@@ -66,7 +62,7 @@ config['signature']: set = { 'id',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_curvature',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
@@ -86,7 +82,7 @@ config['bottleneck_scale'] = 2
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
config['cvae']['input_features'] = 3+3+3+7 if config['use_curvature'] else 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
@@ -112,32 +108,33 @@ config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
####################### cond_ae_curv ############################
|
||||
config['cond_ae_curv'] = dict()
|
||||
config['cond_ae_curv']['device'] = config['device']
|
||||
config['cond_ae_curv']['is_label'] = True
|
||||
config['cond_ae_curv']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_curv']['kernel_size'] = 7
|
||||
config['cond_ae_curv']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
CNBlockConfig(128, config['cond_ae_curv']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_curv']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(config['cond_ae_curv']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
config['cond_ae_curv']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_curv']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_curv']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
@@ -1,197 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '2'
|
||||
config['name'] = 'abl_iccvw_all_2'
|
||||
config['notes'] = 'fixed collision loss'
|
||||
config['tags'] = ['iccv', 'abl_all', 'torch']
|
||||
config['script'] = 'scripts/cvae.py'
|
||||
|
||||
config['pretrained_model'] = 'checkpoints/iccvw/abl_iccvw_all_1'
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_normals'] = True
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
|
||||
config['use_norm_loss'] = True
|
||||
|
||||
ablation_count = [config['use_skinning'], config['use_normals'], config['use_extra_losses']].count(False)
|
||||
assert ablation_count == 1 or ablation_count == 0
|
||||
|
||||
config['avg_garment_path'] = 'custom_data/avg/train_no_shape_avg_uv.npy' #custom_data/train_no_shape_avg_uv.npy # train_no_shape_avg_uv train_no_shape_one_hot_avg_uv
|
||||
if config['use_skinning']:
|
||||
assert (config['avg_garment_path'] != '')
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
###############
|
||||
config['lr'] = 0.0001# 0.0001
|
||||
###############
|
||||
config['starting_balancing_coeff'] = 0.0001
|
||||
config['balancing_activation_limit'] = 0.01
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['upscale_enhancer'] = ''
|
||||
config['freeze_encoders'] = True
|
||||
config['use_one_hot'] = False
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 256, 3),
|
||||
CNBlockConfig(256, config['cvae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'] * 2, 256, 3),
|
||||
CNBlockConfig(256, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 3, 3),
|
||||
]
|
||||
|
||||
config['cvae']['encoder_block_settings'] = block_setting
|
||||
config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,196 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '4'
|
||||
config['name'] = 'abl_iccvw_all_3'
|
||||
config['notes'] = 'fixed collision loss, train from zero'
|
||||
config['tags'] = ['iccv', 'abl_all', 'torch']
|
||||
config['script'] = 'scripts/cvae.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_normals'] = True
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
|
||||
ablation_count = [config['use_skinning'], config['use_normals'], config['use_extra_losses']].count(False)
|
||||
assert ablation_count == 1 or ablation_count == 0
|
||||
|
||||
config['avg_garment_path'] = 'custom_data/avg/train_no_shape_avg_uv.npy' #custom_data/train_no_shape_avg_uv.npy # train_no_shape_avg_uv train_no_shape_one_hot_avg_uv
|
||||
if config['use_skinning']:
|
||||
assert (config['avg_garment_path'] != '')
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
###############
|
||||
config['lr'] = 0.0001# 0.0001
|
||||
###############
|
||||
config['starting_balancing_coeff'] = 0.0001
|
||||
config['balancing_activation_limit'] = 0.01
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['upscale_enhancer'] = ''
|
||||
config['freeze_encoders'] = True
|
||||
config['use_one_hot'] = False
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 256, 3),
|
||||
CNBlockConfig(256, config['cvae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'] * 2, 256, 3),
|
||||
CNBlockConfig(256, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 3, 3),
|
||||
]
|
||||
|
||||
config['cvae']['encoder_block_settings'] = block_setting
|
||||
config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,196 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '4'
|
||||
config['name'] = 'abl_iccvw_all_4'
|
||||
config['notes'] = 'fixed collision loss, train from zero'
|
||||
config['tags'] = ['iccv', 'abl_all', 'torch']
|
||||
config['script'] = 'scripts/cvae.py'
|
||||
|
||||
config['pretrained_model'] = 'checkpoints/iccvw/abl_iccvw_all_3'
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_normals'] = True
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
|
||||
ablation_count = [config['use_skinning'], config['use_normals'], config['use_extra_losses']].count(False)
|
||||
assert ablation_count == 1 or ablation_count == 0
|
||||
|
||||
config['avg_garment_path'] = 'custom_data/avg/train_no_shape_avg_uv.npy' #custom_data/train_no_shape_avg_uv.npy # train_no_shape_avg_uv train_no_shape_one_hot_avg_uv
|
||||
if config['use_skinning']:
|
||||
assert (config['avg_garment_path'] != '')
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
###############
|
||||
config['lr'] = 0.0001# 0.0001
|
||||
###############
|
||||
config['starting_balancing_coeff'] = 0.0001
|
||||
config['balancing_activation_limit'] = 0.01
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['upscale_enhancer'] = ''
|
||||
config['freeze_encoders'] = True
|
||||
config['use_one_hot'] = False
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 256, 3),
|
||||
CNBlockConfig(256, config['cvae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'] * 2, 256, 3),
|
||||
CNBlockConfig(256, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 3, 3),
|
||||
]
|
||||
|
||||
config['cvae']['encoder_block_settings'] = block_setting
|
||||
config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,196 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '4'
|
||||
config['name'] = 'abl_iccvw_all_5'
|
||||
config['notes'] = 'fixed collision loss, train from zero, fixed latest checkpoint loading'
|
||||
config['tags'] = ['iccv', 'abl_all', 'torch']
|
||||
config['script'] = 'scripts/cvae.py'
|
||||
|
||||
config['pretrained_model'] = 'checkpoints/iccvw/abl_iccvw_all_3'
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_normals'] = True
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
|
||||
ablation_count = [config['use_skinning'], config['use_normals'], config['use_extra_losses']].count(False)
|
||||
assert ablation_count == 1 or ablation_count == 0
|
||||
|
||||
config['avg_garment_path'] = 'custom_data/avg/train_no_shape_avg_uv.npy' #custom_data/train_no_shape_avg_uv.npy # train_no_shape_avg_uv train_no_shape_one_hot_avg_uv
|
||||
if config['use_skinning']:
|
||||
assert (config['avg_garment_path'] != '')
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
###############
|
||||
config['lr'] = 0.0001# 0.0001
|
||||
###############
|
||||
config['starting_balancing_coeff'] = 0.0001
|
||||
config['balancing_activation_limit'] = 0.01
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['upscale_enhancer'] = ''
|
||||
config['freeze_encoders'] = True
|
||||
config['use_one_hot'] = False
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 256, 3),
|
||||
CNBlockConfig(256, config['cvae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'] * 2, 256, 3),
|
||||
CNBlockConfig(256, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 3, 3),
|
||||
]
|
||||
|
||||
config['cvae']['encoder_block_settings'] = block_setting
|
||||
config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,196 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'abl_iccvw_all_5_25p'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['iccv', 'abl_all', 'torch']
|
||||
config['script'] = 'scripts/cvae.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_normals'] = True
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
|
||||
ablation_count = [config['use_skinning'], config['use_normals'], config['use_extra_losses']].count(False)
|
||||
assert ablation_count == 1 or ablation_count == 0
|
||||
|
||||
config['avg_garment_path'] = 'custom_data/avg/train_no_shape_avg_uv.npy' #custom_data/train_no_shape_avg_uv.npy # train_no_shape_avg_uv train_no_shape_one_hot_avg_uv
|
||||
if config['use_skinning']:
|
||||
assert (config['avg_garment_path'] != '')
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt_25p.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train_25p.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4_25p'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_25p'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_25p'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
###############
|
||||
config['lr'] = 0.0001# 0.0001
|
||||
###############
|
||||
config['starting_balancing_coeff'] = 0.0001
|
||||
config['balancing_activation_limit'] = 0.01
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['upscale_enhancer'] = ''
|
||||
config['freeze_encoders'] = True
|
||||
config['use_one_hot'] = False
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 256, 3),
|
||||
CNBlockConfig(256, config['cvae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'] * 2, 256, 3),
|
||||
CNBlockConfig(256, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 3, 3),
|
||||
]
|
||||
|
||||
config['cvae']['encoder_block_settings'] = block_setting
|
||||
config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,196 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '1'
|
||||
config['name'] = 'abl_iccvw_all_5_50p'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['iccv', 'abl_all', 'torch']
|
||||
config['script'] = 'scripts/cvae.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_normals'] = True
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
|
||||
ablation_count = [config['use_skinning'], config['use_normals'], config['use_extra_losses']].count(False)
|
||||
assert ablation_count == 1 or ablation_count == 0
|
||||
|
||||
config['avg_garment_path'] = 'custom_data/avg/train_no_shape_avg_uv.npy' #custom_data/train_no_shape_avg_uv.npy # train_no_shape_avg_uv train_no_shape_one_hot_avg_uv
|
||||
if config['use_skinning']:
|
||||
assert (config['avg_garment_path'] != '')
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt_50p.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train_50p.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4_50p'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_50p'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_50p'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
###############
|
||||
config['lr'] = 0.0001# 0.0001
|
||||
###############
|
||||
config['starting_balancing_coeff'] = 0.0001
|
||||
config['balancing_activation_limit'] = 0.01
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['upscale_enhancer'] = ''
|
||||
config['freeze_encoders'] = True
|
||||
config['use_one_hot'] = False
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 256, 3),
|
||||
CNBlockConfig(256, config['cvae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'] * 2, 256, 3),
|
||||
CNBlockConfig(256, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 3, 3),
|
||||
]
|
||||
|
||||
config['cvae']['encoder_block_settings'] = block_setting
|
||||
config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,197 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '3'
|
||||
config['name'] = 'abl_iccvw_all_5_var'
|
||||
config['notes'] = 'fixed collision loss, train from zero, fixed latest checkpoint loading'
|
||||
config['tags'] = ['iccv', 'abl_all', 'torch']
|
||||
config['script'] = 'scripts/cvae.py'
|
||||
|
||||
config['pretrained_model'] = 'checkpoints/iccvw/abl_iccvw_all_5'
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_normals'] = True
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
|
||||
ablation_count = [config['use_skinning'], config['use_normals'], config['use_extra_losses']].count(False)
|
||||
assert ablation_count == 1 or ablation_count == 0
|
||||
|
||||
config['avg_garment_path'] = 'custom_data/avg/train_no_shape_avg_uv.npy' #custom_data/train_no_shape_avg_uv.npy # train_no_shape_avg_uv train_no_shape_one_hot_avg_uv
|
||||
if config['use_skinning']:
|
||||
assert (config['avg_garment_path'] != '')
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
###############
|
||||
config['lr'] = 0.0001# 0.0001
|
||||
config['ignore_saved_optimizer'] = True
|
||||
###############
|
||||
config['starting_balancing_coeff'] = 1
|
||||
config['balancing_activation_limit'] = 0.015
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['upscale_enhancer'] = ''
|
||||
config['freeze_encoders'] = True
|
||||
config['use_one_hot'] = False
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 256, 3),
|
||||
CNBlockConfig(256, config['cvae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'] * 2, 256, 3),
|
||||
CNBlockConfig(256, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 3, 3),
|
||||
]
|
||||
|
||||
config['cvae']['encoder_block_settings'] = block_setting
|
||||
config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,196 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '2'
|
||||
config['name'] = 'abl_iccvw_all_skirt'
|
||||
config['notes'] = 'fixed collision loss, train from zero'
|
||||
config['tags'] = ['iccv', 'abl_all', 'torch']
|
||||
config['script'] = 'scripts/cvae.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_normals'] = True
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
|
||||
ablation_count = [config['use_skinning'], config['use_normals'], config['use_extra_losses']].count(False)
|
||||
assert ablation_count == 1 or ablation_count == 0
|
||||
|
||||
config['avg_garment_path'] = 'custom_data/avg/train_skirt_no_shape_avg_uv.npy' #custom_data/train_no_shape_avg_uv.npy # train_no_shape_avg_uv train_no_shape_one_hot_avg_uv
|
||||
if config['use_skinning']:
|
||||
assert (config['avg_garment_path'] != '')
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = True
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
###############
|
||||
config['lr'] = 0.0001# 0.0001
|
||||
###############
|
||||
config['starting_balancing_coeff'] = 0.0001
|
||||
config['balancing_activation_limit'] = 0.01
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['upscale_enhancer'] = ''
|
||||
config['freeze_encoders'] = True
|
||||
config['use_one_hot'] = False
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 256, 3),
|
||||
CNBlockConfig(256, config['cvae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'] * 2, 256, 3),
|
||||
CNBlockConfig(256, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 3, 3),
|
||||
]
|
||||
|
||||
config['cvae']['encoder_block_settings'] = block_setting
|
||||
config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,197 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '1'
|
||||
config['name'] = 'abl_iccvw_all_skirt_var_2'
|
||||
config['notes'] = 'fixed collision loss, train from zero'
|
||||
config['tags'] = ['iccv', 'abl_all', 'torch']
|
||||
config['script'] = 'scripts/cvae.py'
|
||||
|
||||
config['pretrained_model'] = 'checkpoints/iccvw/abl_iccvw_all_skirt_var'
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_normals'] = True
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
|
||||
ablation_count = [config['use_skinning'], config['use_normals'], config['use_extra_losses']].count(False)
|
||||
assert ablation_count == 1 or ablation_count == 0
|
||||
|
||||
config['avg_garment_path'] = 'custom_data/avg/train_skirt_no_shape_avg_uv.npy' #custom_data/train_no_shape_avg_uv.npy # train_no_shape_avg_uv train_no_shape_one_hot_avg_uv
|
||||
if config['use_skinning']:
|
||||
assert (config['avg_garment_path'] != '')
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = True
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
###############
|
||||
config['lr'] = 0.0001# 0.0001
|
||||
config['ignore_saved_optimizer'] = True
|
||||
###############
|
||||
config['starting_balancing_coeff'] = 2
|
||||
config['balancing_activation_limit'] = 0.015
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['upscale_enhancer'] = ''
|
||||
config['freeze_encoders'] = True
|
||||
config['use_one_hot'] = False
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 256, 3),
|
||||
CNBlockConfig(256, config['cvae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'] * 2, 256, 3),
|
||||
CNBlockConfig(256, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 3, 3),
|
||||
]
|
||||
|
||||
config['cvae']['encoder_block_settings'] = block_setting
|
||||
config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,78 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '2'
|
||||
config['name'] = 'body_iccvw_1'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'body']
|
||||
config['script'] = 'scripts/ae_body_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_mask',
|
||||
'mesh_body_posed',
|
||||
}
|
||||
@@ -1,78 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '5'
|
||||
config['name'] = 'body_iccvw_1_10p'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'body']
|
||||
config['script'] = 'scripts/ae_body_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt_10p.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train_10p.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_mask',
|
||||
'mesh_body_posed',
|
||||
}
|
||||
@@ -1,78 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '5'
|
||||
config['name'] = 'body_iccvw_1_25p'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'body']
|
||||
config['script'] = 'scripts/ae_body_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt_25p.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train_25p.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_mask',
|
||||
'mesh_body_posed',
|
||||
}
|
||||
@@ -1,78 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '5'
|
||||
config['name'] = 'body_iccvw_1_50p'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'body']
|
||||
config['script'] = 'scripts/ae_body_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt_50p.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train_50p.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_mask',
|
||||
'mesh_body_posed',
|
||||
}
|
||||
@@ -1,78 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '1'
|
||||
config['name'] = 'body_iccvw_1_skirt'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'body']
|
||||
config['script'] = 'scripts/ae_body_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = True
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_mask',
|
||||
'mesh_body_posed',
|
||||
}
|
||||
@@ -1,73 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'body_iccvw_cape_f'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'body']
|
||||
config['script'] = 'scripts/ae_body_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/128/cape_processed/train/female'
|
||||
config['valid_dir'] = '/data-local/128/cape_processed/test/female'
|
||||
config['test_dir'] = '/data-local/128/cape_processed/test/female'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = True
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_mask',
|
||||
'mesh_body_posed',
|
||||
}
|
||||
@@ -1,73 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'body_torch_cape_m'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'body']
|
||||
config['script'] = 'scripts/ae_body_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/128/cape_processed/train/male'
|
||||
config['valid_dir'] = '/data-local/128/cape_processed/test/male'
|
||||
config['test_dir'] = '/data-local/128/cape_processed/test/male'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = True
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_mask',
|
||||
'mesh_body_posed',
|
||||
}
|
||||
@@ -1,77 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '1'
|
||||
config['name'] = 'norm_iccvw_1'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'norm']
|
||||
config['script'] = 'scripts/ae_norm_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_mask',
|
||||
'uv_normals',
|
||||
'mesh_mask',
|
||||
'mesh_normals',
|
||||
}
|
||||
@@ -1,77 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'norm_iccvw_1_10p'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'norm']
|
||||
config['script'] = 'scripts/ae_norm_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt_10p.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train_10p.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_mask',
|
||||
'uv_normals',
|
||||
'mesh_mask',
|
||||
'mesh_normals',
|
||||
}
|
||||
@@ -1,77 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'norm_iccvw_1_25p'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'norm']
|
||||
config['script'] = 'scripts/ae_norm_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt_25p.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train_25p.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_mask',
|
||||
'uv_normals',
|
||||
'mesh_mask',
|
||||
'mesh_normals',
|
||||
}
|
||||
@@ -1,77 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '6'
|
||||
config['name'] = 'norm_iccvw_1_50p'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'norm']
|
||||
config['script'] = 'scripts/ae_norm_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt_50p.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train_50p.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_mask',
|
||||
'uv_normals',
|
||||
'mesh_mask',
|
||||
'mesh_normals',
|
||||
}
|
||||
@@ -1,78 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '4'
|
||||
config['name'] = 'norm_iccvw_1_skirt'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'norm']
|
||||
config['script'] = 'scripts/ae_norm_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = True
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_mask',
|
||||
'uv_normals',
|
||||
'mesh_mask',
|
||||
'mesh_normals',
|
||||
}
|
||||
@@ -1,73 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '6'
|
||||
config['name'] = 'norm_iccvw_cape_f'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'norm']
|
||||
config['script'] = 'scripts/ae_norm_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/female'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = True
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_mask',
|
||||
'uv_normals',
|
||||
'mesh_mask',
|
||||
'mesh_normals',
|
||||
}
|
||||
@@ -1,74 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '6'
|
||||
config['name'] = 'norm_iccvw_cape_m'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'norm']
|
||||
config['script'] = 'scripts/ae_norm_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/male'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = True
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_mask',
|
||||
'uv_normals',
|
||||
'mesh_mask',
|
||||
'mesh_normals',
|
||||
}
|
||||
@@ -1,78 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '3'
|
||||
config['name'] = 'static_iccvw_4'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'static']
|
||||
config['script'] = 'scripts/ae_static_train.py'
|
||||
|
||||
config['pretrained_model'] = 'checkpoints/iccvw/static_iccvw_3'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0005
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_static',
|
||||
'mesh_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
@@ -1,78 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '6'
|
||||
config['name'] = 'static_iccvw_4_10p'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'static']
|
||||
config['script'] = 'scripts/ae_static_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt_10p.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train_10p.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0005
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_static',
|
||||
'mesh_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
@@ -1,78 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '6'
|
||||
config['name'] = 'static_iccvw_4_25p'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'static']
|
||||
config['script'] = 'scripts/ae_static_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt_25p.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train_25p.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0005
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_static',
|
||||
'mesh_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
@@ -1,78 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '6'
|
||||
config['name'] = 'static_iccvw_4_50p'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'static']
|
||||
config['script'] = 'scripts/ae_static_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt_50p.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train_50p.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0005
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_static',
|
||||
'mesh_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
@@ -1,78 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '3'
|
||||
config['name'] = 'static_iccvw_1_skirt'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'static']
|
||||
config['script'] = 'scripts/ae_static_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = True
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0005
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_static',
|
||||
'mesh_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
@@ -1,78 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '5'
|
||||
config['name'] = 'static_iccvw_5'
|
||||
config['notes'] = 'lower learning rate'
|
||||
config['tags'] = ['torch', 'static']
|
||||
config['script'] = 'scripts/ae_static_train.py'
|
||||
|
||||
config['pretrained_model'] = 'checkpoints/iccvw/static_iccvw_3'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-net/dataset/test_ready'
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.00005
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_static',
|
||||
'mesh_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
@@ -1,73 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '4'
|
||||
config['name'] = 'static_iccvw_cape_f'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'static']
|
||||
config['script'] = 'scripts/ae_static_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/female'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = True
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0005
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_static',
|
||||
'mesh_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
@@ -1,74 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '5'
|
||||
config['name'] = 'static_iccvw_cape_m'
|
||||
config['notes'] = ' '
|
||||
config['tags'] = ['torch', 'static']
|
||||
config['script'] = 'scripts/ae_static_train.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/male'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['test_frames'] = None
|
||||
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = True
|
||||
config['remove_global_orientation'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cond_ae'] = dict()
|
||||
config['cond_ae']['device'] = config['device']
|
||||
config['cond_ae']['is_label'] = False
|
||||
config['cond_ae']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae']['kernel_size'] = 7
|
||||
config['cond_ae']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
# CNBlockConfig(3, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae']['activation'] = 'tanh'
|
||||
|
||||
|
||||
config['lr'] = 0.0005
|
||||
config['cond_ae']['upscale_output'] = False
|
||||
|
||||
config['signature'] = { 'id',
|
||||
'frame_nr',
|
||||
'uv_static',
|
||||
'mesh_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
@@ -1,151 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '4'
|
||||
config['name'] = 'cvae_norm_1'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['iccv', 'cvae_norm', 'torch']
|
||||
config['script'] = 'scripts/cvae_norm.py'
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
###############
|
||||
config['lr'] = 0.0001# 0.0001
|
||||
###############
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.05
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
'mesh_normals',
|
||||
}
|
||||
|
||||
|
||||
|
||||
config['upscale_enhancer'] = ''
|
||||
config['freeze_encoders'] = True
|
||||
config['use_one_hot'] = False
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 256, 3),
|
||||
CNBlockConfig(256, config['cvae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(176, 256, 3),
|
||||
CNBlockConfig(256, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 3, 3),
|
||||
]
|
||||
|
||||
config['cvae']['encoder_block_settings'] = block_setting
|
||||
config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,152 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '6'
|
||||
config['name'] = 'cvae_norm_2'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['iccv', 'cvae_norm', 'torch']
|
||||
config['script'] = 'scripts/cvae_norm.py'
|
||||
|
||||
config['pretrained_model'] = 'checkpoints/iccvw/cvae_norm_1'
|
||||
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
###############
|
||||
config['lr'] = 0.00003# 0.0001
|
||||
###############
|
||||
config['starting_balancing_coeff'] = 0.1
|
||||
config['balancing_activation_limit'] = 0.1
|
||||
config['ignore_saved_optimizer'] = True
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
'mesh_normals',
|
||||
}
|
||||
|
||||
|
||||
|
||||
config['upscale_enhancer'] = ''
|
||||
config['freeze_encoders'] = True
|
||||
config['use_one_hot'] = False
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 256, 3),
|
||||
CNBlockConfig(256, config['cvae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(176, 256, 3),
|
||||
CNBlockConfig(256, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 3, 3),
|
||||
]
|
||||
|
||||
config['cvae']['encoder_block_settings'] = block_setting
|
||||
config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,152 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '2'
|
||||
config['name'] = 'cvae_norm_3'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['iccv', 'cvae_norm', 'torch']
|
||||
config['script'] = 'scripts/cvae_norm.py'
|
||||
|
||||
config['pretrained_model'] = 'checkpoints/iccvw/cvae_norm_2'
|
||||
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
###############
|
||||
config['lr'] = 0.0001# 0.0001
|
||||
###############
|
||||
config['starting_balancing_coeff'] = 1
|
||||
config['balancing_activation_limit'] = 0.1
|
||||
config['ignore_saved_optimizer'] = True
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
'mesh_normals',
|
||||
}
|
||||
|
||||
|
||||
|
||||
config['upscale_enhancer'] = ''
|
||||
config['freeze_encoders'] = True
|
||||
config['use_one_hot'] = False
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 256, 3),
|
||||
CNBlockConfig(256, config['cvae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(176, 256, 3),
|
||||
CNBlockConfig(256, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 3, 3),
|
||||
]
|
||||
|
||||
config['cvae']['encoder_block_settings'] = block_setting
|
||||
config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,152 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '1'
|
||||
config['name'] = 'cvae_norm_4'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['iccv', 'cvae_norm', 'torch']
|
||||
config['script'] = 'scripts/cvae_norm.py'
|
||||
|
||||
config['pretrained_model'] = 'checkpoints/iccvw/cvae_norm_3'
|
||||
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
###############
|
||||
config['lr'] = 0.0001# 0.0001
|
||||
###############
|
||||
config['starting_balancing_coeff'] = 10
|
||||
config['balancing_activation_limit'] = 0.15
|
||||
config['ignore_saved_optimizer'] = True
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
'mesh_normals',
|
||||
}
|
||||
|
||||
|
||||
|
||||
config['upscale_enhancer'] = ''
|
||||
config['freeze_encoders'] = True
|
||||
config['use_one_hot'] = False
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 256, 3),
|
||||
CNBlockConfig(256, config['cvae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(176, 256, 3),
|
||||
CNBlockConfig(256, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 3, 3),
|
||||
]
|
||||
|
||||
config['cvae']['encoder_block_settings'] = block_setting
|
||||
config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,152 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '1'
|
||||
config['name'] = 'cvae_norm_5'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['iccv', 'cvae_norm', 'torch']
|
||||
config['script'] = 'scripts/cvae_norm.py'
|
||||
|
||||
config['pretrained_model'] = 'checkpoints/iccvw/cvae_norm_4'
|
||||
|
||||
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['resolution'] = 512
|
||||
config['is_skirt'] = False
|
||||
config['dataset_in_memory'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['latent_dim'] = 128
|
||||
|
||||
###############
|
||||
config['lr'] = 0.0001# 0.0001
|
||||
###############
|
||||
config['starting_balancing_coeff'] = 1000
|
||||
config['balancing_activation_limit'] = 0.15
|
||||
config['ignore_saved_optimizer'] = True
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
'mesh_normals',
|
||||
}
|
||||
|
||||
|
||||
|
||||
config['upscale_enhancer'] = ''
|
||||
config['freeze_encoders'] = True
|
||||
config['use_one_hot'] = False
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['cvae'] = dict()
|
||||
config['cvae']['device'] = config['device']
|
||||
config['cvae']['is_label'] = False
|
||||
config['cvae']['input_features'] = 3+3+3
|
||||
config['cvae']['kernel_size'] = 7
|
||||
config['cvae']['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
config['cvae']['latent_dim'] = config['cvae']['bottleneck_size']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 256, 3),
|
||||
CNBlockConfig(256, config['cvae']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cvae']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(176, 256, 3),
|
||||
CNBlockConfig(256, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 3, 3),
|
||||
]
|
||||
|
||||
config['cvae']['encoder_block_settings'] = block_setting
|
||||
config['cvae']['decoder_block_settings'] = block_setting_reverse
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,302 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '1'
|
||||
config['name'] = 'pyr_iccvw_32'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [32]
|
||||
config['pyramid']['pretrained_pyramid'] = ''
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+7 if config['use_curvature'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+7 if config['use_curvature'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+7 if config['use_curvature'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+7 if config['use_curvature'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,302 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '3'
|
||||
config['name'] = 'pyr_iccvw_64'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64]
|
||||
config['pyramid']['pretrained_pyramid'] = ''
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,302 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '2'
|
||||
config['name'] = 'pyr_iccvw_64_128'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_2'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,302 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '2'
|
||||
config['name'] = 'pyr_iccvw_64_128_256'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = '225000'
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,302 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '2'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_256'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = '125000'
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,38 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
|
||||
|
||||
for res in [64, 128, 256]:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_posed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
@@ -1,297 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_cape_f'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_256_cape_f'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = '125000'
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/female'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['test_frames'] = None
|
||||
config['test_frames'] = None
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_cape_f'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_cape_f'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_cape_f'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,297 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '3'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_cape_f_e2e_fine_3'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid_fine.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = ''
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/female'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['test_frames'] = None
|
||||
config['test_frames'] = None
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_cape_f'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_cape_f'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_cape_f'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,297 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '5'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_cape_f_fine_3'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_256_cape_f'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = '125000'
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/female'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['test_frames'] = None
|
||||
config['test_frames'] = None
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_cape_f'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_cape_f'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_cape_f'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,297 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '5'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_cape_f_fine_4'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid_fine.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_256_cape_f'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = '125000'
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/female'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['test_frames'] = None
|
||||
config['test_frames'] = None
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_cape_f'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_cape_f'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_cape_f'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,296 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_cape_m'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_256_cape_m'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = '100000'
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/male'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['test_frames'] = None
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_cape_m'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_torch_cape_m'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_cape_m'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,296 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '2'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_cape_m_e2e'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid_e2e.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = ''
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/male'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['test_frames'] = None
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_cape_m'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_torch_cape_m'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_cape_m'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,296 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '3'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_cape_m_e2e_2'
|
||||
config['notes'] = 'higher learning rate'
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid_e2e.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = ''
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/male'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['test_frames'] = None
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_cape_m'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_torch_cape_m'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_cape_m'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,296 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_cape_m_fine_3'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_256_cape_m'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = '100000'
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/male'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['test_frames'] = None
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_cape_m'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_torch_cape_m'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_cape_m'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,296 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '4'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_cape_m_fine_4'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid_fine.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = 'checkpoints/iccvw/pyr_iccvw_64_128_256_512_cape_m'
|
||||
config['pretrained_model_iteration'] = '25000'
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_256_cape_m'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = '100000'
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/male'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['test_frames'] = None
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_cape_m'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_torch_cape_m'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_cape_m'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,302 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_e2e_2'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid_e2e.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = 'checkpoints/iccvw/pyr_iccvw_64_128_256_512_e2e'
|
||||
config['pretrained_model_iteration'] = '20000'
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = ''
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.01
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,288 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '4'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_norm'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = False
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_256_norm'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
raise NotImplementedError
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,288 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '5'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_norm_skirt'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = True
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = False
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_256_norm_skirt'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
raise NotImplementedError
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,306 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '6'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_reg'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = False # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_256_reg'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,306 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '3'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_reg_e2e'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid_e2e.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = False # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = ''
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,306 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '2'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_reg_skirt'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = True
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = False # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_256_reg_skirt'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,306 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '2'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_reg_skirt_e2e'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid_e2e.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = True
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = False # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = ''
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,309 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '1'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_skin'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = False # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_256_skin'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
else:
|
||||
config['signature'].add('uv_posed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,309 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '2'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_skin_e2e'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid_e2e.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = False # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = ''
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
else:
|
||||
config['signature'].add('uv_posed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,309 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '4'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_skin_skirt'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = True
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = False # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_256_skin_skirt'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
else:
|
||||
config['signature'].add('uv_posed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,309 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '3'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_skin_skirt_e2e'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid_e2e.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = True
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = False # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = ''
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
else:
|
||||
config['signature'].add('uv_posed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,302 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '3'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_skirt'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = True
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_246_skirt'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = '225000'
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,302 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '4'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_512_skirt_e2e'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid_e2e.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = True
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256, 512]
|
||||
config['pyramid']['pretrained_pyramid'] = ''
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,297 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '1'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_cape_f'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_cape_f'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = '100000'
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/female'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['test_frames'] = None
|
||||
config['test_frames'] = None
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_cape_f'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_cape_f'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_cape_f'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,296 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_cape_m'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_cape_m'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = '125000'
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/male'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['test_frames'] = None
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_cape_m'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_torch_cape_m'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_cape_m'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,288 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '6'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_norm'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = False
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_norm'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
raise NotImplementedError
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,288 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '5'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_norm_skirt'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = True
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = False
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_norm_skirt'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
raise NotImplementedError
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,306 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '5'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_reg'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = False # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_reg'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,306 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '6'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_reg_skirt'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = True
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = False # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_reg_skirt'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,309 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '4'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_skin'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = False # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_skin'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
else:
|
||||
config['signature'].add('uv_posed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,309 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '1'
|
||||
config['name'] = 'pyr_iccvw_64_128_256_skin_skirt'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = True
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = False # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_skin_skirt'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
else:
|
||||
config['signature'].add('uv_posed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,302 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '3'
|
||||
config['name'] = 'pyr_iccvw_64_128_246_skirt'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = True
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128, 256]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_128_skirt'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = '300000'
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,297 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '2'
|
||||
config['name'] = 'pyr_iccvw_64_128_cape_f'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_cape_f'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/female'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['test_frames'] = None
|
||||
config['test_frames'] = None
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_cape_f'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_cape_f'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_cape_f'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,297 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config_1 = dict()
|
||||
|
||||
config_1['gpu'] = '2'
|
||||
config_1['name'] = 'pyr_iccvw_64_128_cape_f_2'
|
||||
config_1['notes'] = 'add a scaling to the new level'
|
||||
config_1['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config_1['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config_1['batch_size'] = 4
|
||||
config_1['is_skirt'] = False
|
||||
config_1['is_cape'] = True
|
||||
config_1['device'] = 'cuda'
|
||||
|
||||
|
||||
config_1['use_skinning'] = True # and deshaping
|
||||
config_1['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config_1['use_norm_loss'] = True
|
||||
config_1['use_normals'] = True
|
||||
|
||||
config_1['pretrained_model'] = ''
|
||||
config_1['pretrained_model_iteration'] = ''
|
||||
|
||||
config_1['pyramid'] = dict()
|
||||
config_1['pyramid']['resolutions'] = [64, 128]
|
||||
config_1['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_cape_f'
|
||||
config_1['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config_1['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config_1['train_dir'] = '/data-local/127_2/cape_processed/train/female'
|
||||
config_1['valid_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config_1['test_dir'] = '/data-local/127_2/cape_processed/test/female'
|
||||
config_1['train_frames'] = None
|
||||
config_1['valid_frames'] = None
|
||||
config_1['test_frames'] = None
|
||||
|
||||
config_1['dataset_in_memory'] = False
|
||||
config_1['resolution'] = None
|
||||
|
||||
config_1['smpl_res'] = 14475
|
||||
if config_1['is_cape']:
|
||||
config_1['smpl_res'] = 3638
|
||||
|
||||
config_1['lr'] = 0.0001
|
||||
config_1['starting_balancing_coeff'] = 0.001
|
||||
config_1['balancing_activation_limit'] = 0.012
|
||||
config_1['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config_1['avg_garment_path'] = ''
|
||||
config_1['bottleneck_scale'] = 2
|
||||
|
||||
config_1['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config_1['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config_1['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config_1['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config_1['signature'].add('uv_body_posed' + res_modifier)
|
||||
config_1['signature'].add('uv_unposed' + res_modifier)
|
||||
config_1['signature'].add('uv_normals' + res_modifier)
|
||||
config_1['signature'].add('uv_static' + res_modifier)
|
||||
config_1['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config_1['static_model'] = 'checkpoints/iccvw/static_iccvw_cape_f'
|
||||
config_1['body_model'] = 'checkpoints/iccvw/body_iccvw_cape_f'
|
||||
config_1['normals_model'] = 'checkpoints/iccvw/norm_iccvw_cape_f'
|
||||
config_1['freeze_encoders'] = True
|
||||
|
||||
|
||||
config_1['pyramid']['levels'] = []
|
||||
for res in config_1['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config_1['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config_1['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config_1['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config_1['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config_1['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config_1['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['scaling'] = 0.01
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config_1['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config_1['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config_1['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_cape_f'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config_1['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config_1['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config_1['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config_1['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config_1['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config_1['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config_1['cond_ae_norm'] = dict()
|
||||
config_1['cond_ae_norm']['device'] = config_1['device']
|
||||
config_1['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config_1['cond_ae_norm']['kernel_size'] = 7
|
||||
config_1['cond_ae_norm']['bottleneck_size'] = 40 * config_1['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config_1['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config_1['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config_1['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config_1['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config_1['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config_1['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config_1['cond_ae_body'] = dict()
|
||||
config_1['cond_ae_body']['device'] = config_1['device']
|
||||
config_1['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config_1['cond_ae_body']['kernel_size'] = 7
|
||||
config_1['cond_ae_body']['bottleneck_size'] = 16 * config_1['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config_1['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config_1['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config_1['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config_1['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config_1['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config_1['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config_1['cond_ae_static'] = dict()
|
||||
config_1['cond_ae_static']['device'] = config_1['device']
|
||||
config_1['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config_1['cond_ae_static']['kernel_size'] = 7
|
||||
config_1['cond_ae_static']['bottleneck_size'] = 8 * config_1['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config_1['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config_1['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config_1['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config_1['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config_1['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config_1['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,296 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'pyr_iccvw_64_128_cape_m'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_cape_m'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/male'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['test_frames'] = None
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_cape_m'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_torch_cape_m'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_cape_m'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,297 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '3'
|
||||
config['name'] = 'pyr_iccvw_64_128_cape_m_2'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = True
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_cape_m'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127_2/cape_processed/train/male'
|
||||
config['valid_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['test_dir'] = '/data-local/127_2/cape_processed/test/male'
|
||||
config['train_frames'] = None
|
||||
config['valid_frames'] = None
|
||||
config['test_frames'] = None
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'mesh_unposed',
|
||||
'uv_unposed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_cape_m'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_torch_cape_m'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_cape_m'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['scaling'] = 0.01
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_cape_m'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,288 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '5'
|
||||
config['name'] = 'pyr_iccvw_64_128_norm'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = False
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_norm'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
raise NotImplementedError
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,288 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '4'
|
||||
config['name'] = 'pyr_iccvw_64_128_norm_skirt'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = True
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = False
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_norm_skirt_2'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2 - 40*2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
raise NotImplementedError
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,306 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '1'
|
||||
config['name'] = 'pyr_iccvw_64_128_reg'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = False # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_reg'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,306 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '6'
|
||||
config['name'] = 'pyr_iccvw_64_128_reg_skirt'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = True
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = True # and deshaping
|
||||
config['use_extra_losses'] = False # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_reg_skirt_3'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,309 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '5'
|
||||
config['name'] = 'pyr_iccvw_64_128_skin'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = False
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = False # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_skin'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
else:
|
||||
config['signature'].add('uv_posed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_4'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
@@ -1,309 +0,0 @@
|
||||
from utils.convnext_utils import CNBlockConfig
|
||||
|
||||
config = dict()
|
||||
|
||||
config['gpu'] = '0'
|
||||
config['name'] = 'pyr_iccvw_64_128_skin_skirt'
|
||||
config['notes'] = ''
|
||||
config['tags'] = ['pyr', 'iccv', 'torch']
|
||||
config['script'] = 'scripts/cvae_pyramid.py'
|
||||
|
||||
config['batch_size'] = 4
|
||||
config['is_skirt'] = True
|
||||
config['is_cape'] = False
|
||||
config['device'] = 'cuda'
|
||||
|
||||
|
||||
config['use_skinning'] = False # and deshaping
|
||||
config['use_extra_losses'] = True # edge, collision, 3d mesh
|
||||
config['use_norm_loss'] = True
|
||||
config['use_normals'] = True
|
||||
|
||||
config['pretrained_model'] = ''
|
||||
config['pretrained_model_iteration'] = ''
|
||||
|
||||
config['pyramid'] = dict()
|
||||
config['pyramid']['resolutions'] = [64, 128]
|
||||
config['pyramid']['pretrained_pyramid'] = 'checkpoints/iccvw/pyr_iccvw_64_skin_skirt_3'
|
||||
config['pyramid']['pretrained_pyramid_iteration'] = ''
|
||||
config['pyramid']['final_upscale_enhancer'] = ''
|
||||
|
||||
config['train_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['valid_dir'] = '/data-local/127/cloth3d_processed'
|
||||
config['test_dir'] = '/data-local/127_2/cloth3d_processed_test'
|
||||
|
||||
if config['is_skirt']:
|
||||
config['train_frames'] = 'frames_train_skirt.txt'
|
||||
config['valid_frames'] = 'frames_valid_skirt.txt'
|
||||
config['test_frames'] = 'frames_test_skirt.txt'
|
||||
else:
|
||||
config['train_frames'] = 'frames_train.txt'
|
||||
config['valid_frames'] = 'frames_valid.txt'
|
||||
config['test_frames'] = 'frames_test.txt'
|
||||
|
||||
config['dataset_in_memory'] = False
|
||||
config['resolution'] = None
|
||||
|
||||
config['smpl_res'] = 14475
|
||||
if config['is_cape']:
|
||||
config['smpl_res'] = 3638
|
||||
|
||||
config['lr'] = 0.0001
|
||||
config['starting_balancing_coeff'] = 0.001
|
||||
config['balancing_activation_limit'] = 0.012
|
||||
config['reg_loss_coeff'] = 0.1 # set to 0 if don't want to use edge+collision
|
||||
config['avg_garment_path'] = ''
|
||||
config['bottleneck_scale'] = 2
|
||||
|
||||
config['signature']: set = { 'id',
|
||||
'frame_nr',
|
||||
'pose_shape',
|
||||
'uv_body_posed',
|
||||
'mesh_body_posed',
|
||||
'mesh_posed',
|
||||
'uv_normals',
|
||||
'uv_static',
|
||||
'mesh_mask',
|
||||
'uv_mask',
|
||||
}
|
||||
|
||||
config['pyramid']['is_label'] = False
|
||||
# config['pyramid']['add_noise'] = False
|
||||
# config['pyramid']['noise_amount'] = 0.0
|
||||
|
||||
config['pyramid']['loss_weights'] = [1,1,1,1,1]
|
||||
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed')
|
||||
config['signature'].add('mesh_unposed')
|
||||
else:
|
||||
config['signature'].add('uv_posed')
|
||||
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res != 512:
|
||||
res_modifier = '_' + str(res)
|
||||
config['signature'].add('uv_body_posed' + res_modifier)
|
||||
if config['use_skinning']:
|
||||
config['signature'].add('uv_unposed' + res_modifier)
|
||||
else:
|
||||
config['signature'].add('uv_posed' + res_modifier)
|
||||
config['signature'].add('uv_normals' + res_modifier)
|
||||
config['signature'].add('uv_static' + res_modifier)
|
||||
config['signature'].add('uv_mask' + res_modifier)
|
||||
|
||||
|
||||
config['static_model'] = 'checkpoints/iccvw/static_iccvw_1_skirt'
|
||||
config['body_model'] = 'checkpoints/iccvw/body_iccvw_1_skirt'
|
||||
config['normals_model'] = 'checkpoints/iccvw/norm_iccvw_1_skirt'
|
||||
config['freeze_encoders'] = True
|
||||
|
||||
|
||||
config['pyramid']['levels'] = []
|
||||
for res in config['pyramid']['resolutions']:
|
||||
if res == 512:
|
||||
level = dict()
|
||||
level['resolution'] = 512
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['kernel_size'] = 7
|
||||
level['upscale_enhancer'] = ''
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 3),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 256:
|
||||
level = dict()
|
||||
level['resolution'] = 256
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_256_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 3),
|
||||
CNBlockConfig(128, 128, 3),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 3),
|
||||
CNBlockConfig(128, 128, 9),
|
||||
CNBlockConfig(128, 64, 3),
|
||||
CNBlockConfig(64, 3, 3),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 128:
|
||||
level = dict()
|
||||
level['resolution'] = 128
|
||||
level['kernel_size'] = 5
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_128_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 128, 9),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 64:
|
||||
level = dict()
|
||||
level['resolution'] = 64
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_64_skirt'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(64, level['bottleneck_size'], 9),
|
||||
CNBlockConfig(level['bottleneck_size'], None, 9),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 64, 9),
|
||||
CNBlockConfig(64, 3, 9),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
elif res == 32:
|
||||
level = dict()
|
||||
level['resolution'] = 32
|
||||
level['kernel_size'] = 3
|
||||
level['upscale_enhancer'] = 'checkpoints/iccvw/iccv_32'
|
||||
level['upscale_enhancer_freeze'] = True
|
||||
level['bottleneck_size'] = 64 * config['bottleneck_scale']
|
||||
level['input_features'] = 3+3+3+3 if config['use_normals'] else 3+3+3
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(level['bottleneck_size'], None, 16),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(level['bottleneck_size'] * 2, 3, 16),
|
||||
]
|
||||
|
||||
level['encoder_block_settings'] = block_setting
|
||||
level['decoder_block_settings'] = block_setting_reverse
|
||||
config['pyramid']['levels'].append(level)
|
||||
else:
|
||||
raise NotImplementedError
|
||||
|
||||
|
||||
####################### cond_ae_norm ############################
|
||||
config['cond_ae_norm'] = dict()
|
||||
config['cond_ae_norm']['device'] = config['device']
|
||||
config['cond_ae_norm']['is_label'] = False
|
||||
# config['cond_ae_norm']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_norm']['kernel_size'] = 7
|
||||
config['cond_ae_norm']['bottleneck_size'] = 40 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_norm']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_norm']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_norm']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_norm']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_norm']['activation'] = 'tanh'
|
||||
|
||||
|
||||
####################### cond_ae_body ############################
|
||||
config['cond_ae_body'] = dict()
|
||||
config['cond_ae_body']['device'] = config['device']
|
||||
config['cond_ae_body']['is_label'] = False
|
||||
# config['cond_ae_body']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_body']['kernel_size'] = 7
|
||||
config['cond_ae_body']['bottleneck_size'] = 16 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, 128, 9),
|
||||
CNBlockConfig(128, config['cond_ae_body']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_body']['bottleneck_size'], 128, 3),
|
||||
CNBlockConfig(128, 64, 9),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_body']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_body']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_body']['activation'] = 'tanh'
|
||||
|
||||
####################### cond_ae_static ############################
|
||||
config['cond_ae_static'] = dict()
|
||||
config['cond_ae_static']['device'] = config['device']
|
||||
config['cond_ae_static']['is_label'] = False
|
||||
# config['cond_ae_static']['latent_dim'] = config['latent_dim']
|
||||
config['cond_ae_static']['kernel_size'] = 7
|
||||
config['cond_ae_static']['bottleneck_size'] = 8 * config['bottleneck_scale']
|
||||
|
||||
block_setting = [
|
||||
CNBlockConfig(8, 16, 3),
|
||||
CNBlockConfig(16, 32, 3),
|
||||
CNBlockConfig(32, 64, 3),
|
||||
CNBlockConfig(64, config['cond_ae_static']['bottleneck_size'], 9),
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], None, 3),
|
||||
]
|
||||
|
||||
block_setting_reverse = [
|
||||
CNBlockConfig(config['cond_ae_static']['bottleneck_size'], 64, 3),
|
||||
CNBlockConfig(64, 32, 9),
|
||||
CNBlockConfig(32, 16, 3),
|
||||
CNBlockConfig(16, 8, 3),
|
||||
CNBlockConfig(8, 3, 3),
|
||||
]
|
||||
|
||||
config['cond_ae_static']['encoder_block_settings'] = block_setting
|
||||
config['cond_ae_static']['decoder_block_settings'] = block_setting_reverse
|
||||
config['cond_ae_static']['activation'] = 'tanh'
|
||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user