training fix

aaa_helpers.txt

@@ -24,6 +24,8 @@ nohup python cvae.py --name="lhunor_abl_reg" > nohup/iccv/ablation/abl_reg.out 2
 nohup python cvae.py --name="lhunor_abl_off" > nohup/iccv/ablation/abl_off.out 2>&1 &
 nohup python cvae.py --name="lhunor_abl_skin_curv" > nohup/iccv/ablation/abl_skin_curv.out 2>&1 &
 
+nohup python cvae.py --name="lhunor_abl_skin_gnn" > nohup/iccv/ablation/abl_skin_gnn.out 2>&1 &
+
 nohup python upscale_test.py --name="iccv_32_2" > nohup/iccv/upscale/iccv_32_2.out 2>&1 &
 nohup python upscale_test.py --name="iccv_64_2" > nohup/iccv/upscale/iccv_64_2.out 2>&1 &
 nohup python upscale_test.py --name="iccv_128_2" > nohup/iccv/upscale/iccv_128_2.out 2>&1 &

configs/iccv/config_abl_all.py

@@ -14,6 +14,7 @@ config['use_extra_losses'] =          True # edge, collision, 3d mesh
 config['use_offsets'] =               True
 
 config['use_norm_loss'] =             True
+config['only_curvatures'] =           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
@@ -43,35 +44,38 @@ config['static_model'] = 'checkpoints/iccv/static_iccv_1'
 config['body_model'] = 'checkpoints/iccv/body_iccv_1'
 config['curvature_model'] = 'checkpoints/iccv/curv_iccv_1'
 
+config['static_model'] = 'checkpoints/post/static_post_2'
+config['body_model'] = 'checkpoints/post/body_post_2'
+config['curvature_model'] = 'checkpoints/post/curv_post_1'
+
 
 config['smpl_res'] = 14475
 if config['is_cape']:
   config['smpl_res'] = 3638
 
-config['latent_dim'] = 128
 
 config['pretrained_model'] = ''
 
 ###############    
 config['lr'] = 0.0001#                 0.0001
 ###############
-config['starting_balancing_coeff'] = 0.0001
+config['starting_balancing_coeff'] = 0.001
 
 config['signature_dynamic'] = { 'id': (), 'frame_nr': (),
                           'pose_shape': (72 + 11,),
                           'uv_body_posed': (512, 256, 3),
-                          'mesh_body_posed': (14475, 3,),
+                          'mesh_body_posed': (config['smpl_res'], 3,),
-                          'mesh_posed': (14475, 3,),
+                          'mesh_posed': (config['smpl_res'], 3,),
                           'uv_curvature': (512, 256, 7),
                         }
 config['signature_static'] = { 'uv_static': (512, 256, 3),
-                              'mesh_mask': (14475,),
+                              'mesh_mask': (config['smpl_res'],),
                               'uv_mask': (512, 256,),
                               }
 
 if config['use_skinning']:
   config['signature_dynamic']['uv_unposed'] = (512, 256, 3)
-  config['signature_dynamic']['mesh_unposed'] = (14475, 3,)
+  config['signature_dynamic']['mesh_unposed'] = (config['smpl_res'], 3,)
 else:
   config['signature_dynamic']['uv_posed'] = (512, 256, 3)
 

configs/iccv/config_abl_skin.py

@@ -18,8 +18,8 @@ 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['gpu'] = '2'
+config['gpu'] = '4'
-config['name'] = 'abl_skin'
+config['name'] = 'abl_skin_gnn'
 config['notes'] = ''
 config['tags'] = ['iccv', 'abl_skin']
 

cvae.py

@@ -1,6 +1,6 @@
 import os
 import sys
-from configs.iccv.config_abl_skin_curv import config
+from configs.iccv.config_abl_skin import config
 
 os.environ["CUDA_VISIBLE_DEVICES"] = config['gpu']
 import tensorflow as tf
@@ -34,7 +34,6 @@ else:
     from losses import edge_loss
 
 
-
 from losses import collision_ratio, kl_loss, normal_loss #, concept_loss
 # from losses import collision_penalty as collision_penalty
 from losses import collision_loss as collision_penalty
@@ -42,6 +41,7 @@ from custom_utils.custom_utils import batch_to_dict, load_model, plot_3d, UV2Mes
 from dataset import DatasetFactory, StaticDataset
 from custom_utils.smpl import SMPL
 from models.cvae_model import CVAENew as CVAE
+from models.gnn import Smoother
 from custom_utils.custom_utils import Accumulator
 
 
@@ -54,6 +54,7 @@ def init_wandb():
     # artifact.add_file('configs/config_base.py')
     # artifact.add_file('configs/config_cvae.py')
     artifact.add_file('models/base.py')
+    artifact.add_file('models/gnn.py')
     artifact.add_file('models/cond_encoder.py')
     artifact.add_file('dataset.py')
     artifact.add_file('convnext/models/convnext_tf.py')
@@ -88,6 +89,7 @@ def train_step(batch):
         pred, mean, logvar = model([uv_orig, batch['uv_static'], batch['uv_body_posed'], batch['uv_curvature'], batch['uv_mask']], training=True)
 
         mesh_pred = uv2mesh(pred)
+        mesh_pred = smoother(mesh_pred, training=True)
 
         if config['use_skinning']:
             shape = batch['pose_shape'][:, 72:]
@@ -126,8 +128,8 @@ def train_step(batch):
 
     # ms_ssim = tf.reduce_mean(tf.image.ssim_multiscale(pred + 5, batch['uv_unposed'] + 5, max_val=10, power_factors=(0.5, 0.2, 0.1, 0.1, 0.1)))
     metric = tf.reduce_sum(tf.math.sqrt(tf.reduce_sum(batch['mesh_mask'][:,:,tf.newaxis] * (mesh_orig - mesh_pred) ** 2, axis=-1))) / tf.reduce_sum(batch['mesh_mask'])
-    gradients = tape.gradient(loss, model.trainable_variables)
+    gradients = tape.gradient(loss, model.trainable_variables + smoother.trainable_variables)
-    optimizer.apply_gradients(zip(gradients, model.trainable_variables))
+    optimizer.apply_gradients(zip(gradients, model.trainable_variables + smoother.trainable_variables))
     return { 'train/loss': loss, 'train/edge_loss': e_loss, 'train/collision_loss': c_loss, 'train/euclid_error': metric,
              'train/reconstr_loss': reconstr_loss, 'train/collision_ratio': c_ratio,
              'train/mean': mean, 'train/logvar': logvar, 'train/kl_loss': kl, 'train/var': np.e ** logvar,
@@ -146,6 +148,7 @@ def valid_step(batch):
     pred, mean, logvar = model([uv_orig, batch['uv_static'], batch['uv_body_posed'], batch['uv_curvature'], batch['uv_mask']], training=False)
 
     mesh_pred = uv2mesh(pred)
+    mesh_pred = smoother(mesh_pred, training=False)
 
     if config['use_skinning']:
         shape = batch['pose_shape'][:, 72:]
@@ -255,7 +258,8 @@ def plot_everything(batch: dict, type: str):
 
     pred, mean, logvar = model([uv_orig, batch['uv_static'], batch['uv_body_posed'], batch['uv_curvature'], batch['uv_mask']], training=False)
     
-    mesh_pred = uv2mesh(pred).numpy()
+    mesh_pred = uv2mesh(pred)
+    mesh_pred = smoother(mesh_pred, training=False).numpy()
     
     orig_img = uv_orig
     uv_static_img = batch['uv_static']
@@ -520,6 +524,7 @@ if __name__ == '__main__':
                                         is_cape=config['is_cape'])
 
     model = CVAE(config)
+    smoother = Smoother()
 
     optimizer = tf.optimizers.Adam(learning_rate=config['lr'])
     balancing_activated_step = 0
@@ -546,7 +551,7 @@ if __name__ == '__main__':
         train_loss = train_step(batch)
         train_accumulator.update(train_loss)
 
-        if step % 50 == 0:
+        if step % 25 == 0:
             data_valid = valid_iterator.get_next()
             data_valid = static_dataset_valid.expand_batch(batch_to_dict(data_valid, config['signature_dynamic']))
             valid_loss = valid_step(data_valid)

models/gnn.py

@@ -0,0 +1,40 @@
+import os
+import sys
+
+import scipy.io as sio
+import tensorflow as tf
+from spektral.layers import GCNConv, GeneralConv
+from spektral.utils import sp_matrix_to_sp_tensor
+from tensorflow.keras.layers import (Conv2D, Conv2DTranspose, Dense, DepthwiseConv2D,
+                                     GlobalAveragePooling2D, Layer,
+                                     LayerNormalization, Flatten, Reshape, UpSampling2D)
+
+sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+
+class Smoother(tf.keras.Model):
+    def __init__(self, ):
+        super().__init__()
+        self.a = sp_matrix_to_sp_tensor(sio.loadmat("custom_utils/ssmpl0.mat")['L'].tocoo())
+        self.conv1 = GCNConv(16, activation='tanh', use_bias=False)
+        self.conv2 = GCNConv(3, activation='tanh', use_bias=False)
+        # self.conv3 = GCNConv(3, activation='tanh', use_bias=False)
+        # self.conv4 = GCNConv(3, activation='tanh', use_bias=False)
+        # self.conv5 = GCNConv(3, activation='tanh', use_bias=False)
+        self.conv6 = GCNConv(3, activation='tanh', use_bias=False)
+
+    def call(self, inputs, training=None):
+        x = inputs
+        x = self.conv1([x, self.a], training=training)
+        x = self.conv2([x, self.a], training=training)
+        # x = self.conv3([x, self.a], training=training)
+        # x = self.conv4([x, self.a], training=training)
+        # x = self.conv5([x, self.a], training=training)
+        x = self.conv6([x, self.a], training=training)
+        return x
+
+
+
+
+
+