438 lines
17 KiB
Python
438 lines
17 KiB
Python
import os
|
|
os.environ["CUDA_VISIBLE_DEVICES"] = "1"
|
|
import tensorflow as tf
|
|
|
|
gpus = tf.config.experimental.list_physical_devices('GPU')
|
|
if gpus:
|
|
try:
|
|
# Currently, memory growth needs to be the same across GPUs
|
|
for gpu in gpus:
|
|
tf.config.experimental.set_memory_growth(gpu, True)
|
|
logical_gpus = tf.config.experimental.list_logical_devices('GPU')
|
|
print(len(gpus), "Physical GPUs,", len(logical_gpus), "Logical GPUs")
|
|
except RuntimeError as e:
|
|
# Memory growth must be set before GPUs have been initialized
|
|
print(e)
|
|
|
|
import time
|
|
import argparse
|
|
import wandb
|
|
import random
|
|
import cmapy as cm
|
|
import numpy as np
|
|
import scipy as sp
|
|
from custom_utils.custom_utils import UV2Mesh, UV2MeshLabel, maskFacesBatch, _maskFaces
|
|
import plotly.graph_objects as go
|
|
import plotly.express as ex
|
|
from PIL import Image
|
|
|
|
from tf_utils import check_tf_version, allow_memory_growth
|
|
# from losses import collision_penalty as collision_penalty
|
|
from losses import collision_loss as collision_penalty
|
|
from dataset import DatasetFactory, StaticDataset
|
|
from custom_utils.custom_utils import batch_to_dict
|
|
from custom_utils.smpl import SMPL
|
|
from models.cvae_curv_model import CVAE
|
|
|
|
|
|
name='cvae_curv_3'
|
|
def init_wandb():
|
|
run = wandb.init(project='vae', name=name,
|
|
notes='learn offsets. smaller flattened latent vectors. with 25-85k pretraining. smaller batchsize. new implementation for kl loss. \
|
|
no lr schedule. 25 jumpsuits, average reconst_loss before calculation, offsets over average unposed garment. gamma balancing. remove global \
|
|
orientation (in posed body input and in pose_shape parameters.. r lower kl coeff to 0.1. fixed plotting colors\
|
|
addent MLP for lowering conditional Z diemnsion\
|
|
higher starting kl coeff (0.1)\
|
|
train on 100 garments')
|
|
artifact = wandb.Artifact('code', type='code')
|
|
artifact.add_file('cvae_curv.py')
|
|
artifact.add_file('models/cvae_curv_model.py')
|
|
artifact.add_file('dataset.py')
|
|
artifact.add_file('losses.py')
|
|
artifact.add_file('convnext/models/convnext_tf.py')
|
|
run.log_artifact(artifact)
|
|
|
|
uv2mesh = UV2MeshLabel(res=256)
|
|
|
|
def plot_3d(curv_gt, curv_pred, faces, mesh):
|
|
mesh_2 = mesh.copy()
|
|
mesh_2[:,1] += 1
|
|
|
|
colors = np.array([cm.color('jet', 1 - i) for i in np.linspace(0, 1, 6+1)])
|
|
v_color_1 = np.zeros((mesh.shape[0], 3))
|
|
v_color_2 = np.zeros((mesh.shape[0], 3))
|
|
for i in range(6+1):
|
|
v_color_1[curv_gt == i, :] = colors[i]
|
|
v_color_2[curv_pred == i, :] = colors[i]
|
|
|
|
fig = go.Figure(data=[
|
|
go.Mesh3d(
|
|
# 8 vertices of a cube
|
|
x=mesh[:,0],
|
|
y=mesh[:,1],
|
|
z=mesh[:,2],
|
|
# i, j and k give the vertices of triangles
|
|
i = faces[:,0],
|
|
j = faces[:,1],
|
|
k = faces[:,2],
|
|
vertexcolor=v_color_1,
|
|
text='gt'
|
|
),
|
|
go.Mesh3d(
|
|
# 8 vertices of a cube
|
|
x=mesh_2[:,0],
|
|
y=mesh_2[:,1],
|
|
z=mesh_2[:,2],
|
|
# i, j and k give the vertices of triangles
|
|
i = faces[:,0],
|
|
j = faces[:,1],
|
|
k = faces[:,2],
|
|
vertexcolor=v_color_2,
|
|
text='pred'
|
|
)
|
|
])
|
|
|
|
fig.update_layout(scene_aspectmode='manual',
|
|
scene_aspectratio=dict(x=1, y=2, z=1))
|
|
fig.update_layout(
|
|
scene = dict(
|
|
xaxis = dict(nticks=4, range=[-1,1],),
|
|
yaxis = dict(nticks=4, range=[-1,3],),
|
|
zaxis = dict(nticks=4, range=[-1,1],),)
|
|
)
|
|
|
|
return fig
|
|
|
|
|
|
def log_normal_pdf(sample, mean, logvar, raxis=1):
|
|
log2pi = tf.math.log(2. * np.pi)
|
|
res = tf.reduce_sum(-.5 * ((sample - mean) ** 2. * tf.exp(-logvar) + logvar + log2pi), axis=raxis)
|
|
return res
|
|
|
|
|
|
def kl_loss(mean, logvar):
|
|
kl_loss = 1 + logvar - tf.square(mean) - tf.exp(logvar)
|
|
kl_loss = tf.reduce_mean(kl_loss)
|
|
kl_loss *= -0.5
|
|
return kl_loss
|
|
|
|
|
|
|
|
g_mean_iou = tf.keras.metrics.MeanIoU(num_classes=7)
|
|
def get_mean_iou(y_true, y_pred):
|
|
g_mean_iou.reset_state()
|
|
g_mean_iou.update_state(y_true, y_pred)
|
|
return g_mean_iou.result()
|
|
|
|
# https://github.com/NifTK/NiftyNet/blob/dev/niftynet/layer/loss_segmentation.py
|
|
def dice_dense(prediction, ground_truth, weight_map=None):
|
|
"""
|
|
Computing mean-class Dice similarity.
|
|
:param prediction: last dimension should have ``num_classes``
|
|
:param ground_truth: segmentation ground truth (encoded as a binary matrix)
|
|
last dimension should be ``num_classes``
|
|
:param weight_map:
|
|
:return: ``1.0 - mean(Dice similarity per class)``
|
|
"""
|
|
|
|
if weight_map is not None:
|
|
raise NotImplementedError
|
|
prediction = tf.cast(prediction, dtype=tf.float32)
|
|
ground_truth = tf.cast(ground_truth, dtype=tf.float32)
|
|
ground_truth = tf.reshape(ground_truth, prediction.shape)
|
|
# computing Dice over the spatial dimensions
|
|
reduce_axes = list(range(len(prediction.shape) - 1))
|
|
dice_numerator = 2.0 * tf.reduce_sum(
|
|
prediction * ground_truth, axis=reduce_axes)
|
|
dice_denominator = \
|
|
tf.reduce_sum(tf.square(prediction), axis=reduce_axes) + \
|
|
tf.reduce_sum(tf.square(ground_truth), axis=reduce_axes)
|
|
|
|
epsilon = 0.00001
|
|
|
|
dice_score = (dice_numerator + epsilon) / (dice_denominator + epsilon)
|
|
return tf.reduce_mean(dice_score)
|
|
|
|
|
|
@tf.function
|
|
def train_step(batch):
|
|
with tf.GradientTape() as tape:
|
|
uv_curvature = tf.one_hot(batch['uv_curvature'], 7)
|
|
|
|
pred, mean, logvar, z = model([batch['uv_static'], batch['uv_body_posed'], uv_curvature, batch['uv_mask']])
|
|
|
|
ce_uv_loss = scce(batch['uv_curvature'], pred)
|
|
ms_ssim = tf.reduce_mean(tf.image.ssim(pred, uv_curvature, max_val=1))
|
|
|
|
reconstr_loss = ce_uv_loss
|
|
reconstr_loss = tf.reduce_mean(reconstr_loss)
|
|
kl = kl_loss(mean, logvar)
|
|
loss = reconstr_loss + balancing_coeff * kl
|
|
ms_ssim = tf.reduce_mean(ms_ssim)
|
|
mean = tf.reduce_mean(mean)
|
|
logvar = tf.reduce_mean(logvar)
|
|
|
|
|
|
# mesh_pred = uv2mesh(tf.tile(tf.math.argmax(pred, axis=-1, output_type=tf.dtypes.int32)[:,:,:,tf.newaxis], [1,1,1,3]))[:,:,0]
|
|
# mesh_orig = tf.one_hot(batch['mesh_curvature'] + tf.cast(batch['mesh_mask'], tf.int32), 7)
|
|
# ce_mesh_loss = scce(mesh_orig, mesh_pred)
|
|
|
|
mean_iou = get_mean_iou(batch['uv_curvature'], tf.math.argmax(pred, axis=-1))
|
|
dice = dice_dense(pred, uv_curvature)
|
|
|
|
gradients = tape.gradient(loss, model.trainable_variables)
|
|
optimizer.apply_gradients(zip(gradients, model.trainable_variables))
|
|
return { 'train/loss': loss, 'train/ce_uv_loss': ce_uv_loss, 'train/ssim': ms_ssim, 'train/mean_iou': mean_iou, 'train/dice': dice, #'ce_mesh_loss': ce_mesh_loss
|
|
'train/reconstr_loss': reconstr_loss,
|
|
'train/mean': mean, 'train/logvar': logvar, 'train/kl_loss': kl, 'train/reconst_loss_norm': reconstr_loss / (2 * (gamma ** 2)), 'train/var': np.e ** logvar,
|
|
'train/reg_coeff': reg_coeff}
|
|
|
|
|
|
|
|
|
|
|
|
@tf.function
|
|
def valid_step(batch):
|
|
uv_curvature = tf.one_hot(batch['uv_curvature'], 7)
|
|
|
|
pred, mean, logvar, z = model([batch['uv_static'], batch['uv_body_posed'], uv_curvature, batch['uv_mask']])
|
|
|
|
ce_uv_loss = scce(batch['uv_curvature'], pred)
|
|
ms_ssim = tf.reduce_mean(tf.image.ssim(pred, uv_curvature, max_val=1))
|
|
loss = ce_uv_loss
|
|
|
|
mean_iou = get_mean_iou(batch['uv_curvature'], tf.math.argmax(pred, axis=-1))
|
|
dice = dice_dense(pred, uv_curvature)
|
|
|
|
reconstr_loss = tf.reduce_mean(ce_uv_loss)
|
|
kl = kl_loss(mean, logvar)
|
|
loss = reconstr_loss + balancing_coeff * kl
|
|
ms_ssim = tf.reduce_mean(ms_ssim)
|
|
mean = tf.reduce_mean(mean)
|
|
logvar = tf.reduce_mean(logvar)
|
|
|
|
return { 'valid/loss': loss, 'valid/ce_uv_loss': ce_uv_loss, 'valid/ssim': ms_ssim, 'valid/mean_iou': mean_iou, 'valid/dice': dice,
|
|
'valid/ms_ssim': ms_ssim, 'valid/reconstr_loss': reconstr_loss,
|
|
'valid/mean': mean, 'valid/logvar': logvar, 'valid/kl_loss': kl, 'valid/var': np.e ** logvar }
|
|
|
|
|
|
def plot_everything(batch: dict, type: str):
|
|
uv_curvature = tf.one_hot(batch['uv_curvature'], 7)
|
|
pred, mean, logvar, z = model([batch['uv_static'], batch['uv_body_posed'], uv_curvature, batch['uv_mask']])
|
|
|
|
pred = np.argmax(pred.numpy(), axis=-1)
|
|
mesh_pred = uv2mesh(pred)
|
|
mesh_orig = batch['mesh_curvature']
|
|
mesh_base = batch['mesh_posed']
|
|
|
|
orig_img = batch['uv_curvature']
|
|
pred_img = pred
|
|
img = tf.concat((orig_img, pred_img), axis=2)
|
|
imgs = [wandb.Image(img[i].numpy().astype(np.float32) / 6.0) for i in range(4)]
|
|
|
|
F = np.load("custom_utils/smpl_faces.npy")
|
|
mask = batch['mesh_mask'].numpy().astype(bool)
|
|
_, faces = maskFacesBatch(np.zeros((4, 14475, 3)), F, mask)
|
|
gt_curv = [mesh_orig[i].numpy()[mask[i]] + 1 for i in range(pred.shape[0])]
|
|
pred_curv = [mesh_pred[i].numpy()[mask[i]] for i in range(pred.shape[0])]
|
|
mesh = [mesh_base[i].numpy()[mask[i]] for i in range(pred.shape[0])]
|
|
|
|
figs = { type + "/garment_" + str(i): plot_3d(gt_curv[i], pred_curv[i], faces[i], mesh[i]) for i in range(4) }
|
|
|
|
mean_hist = wandb.Histogram(sequence=mean.numpy().mean(axis=0), num_bins=model.latent_dim)
|
|
logvar_hist = wandb.Histogram(sequence=logvar.numpy().mean(axis=0), num_bins=model.latent_dim)
|
|
|
|
return { **figs, **{ type + '/images': imgs, type + '/mean_hist': mean_hist, type + '/logvar_hist': logvar_hist } }
|
|
|
|
|
|
def load_model(model, checkpoint):
|
|
ckpt = tf.train.Checkpoint(model=model)
|
|
manager = tf.train.CheckpointManager(ckpt, checkpoint, max_to_keep=1)
|
|
ckpt.restore(manager.latest_checkpoint)
|
|
|
|
|
|
def sample(folder):
|
|
load_model(model,folder)
|
|
for batch in dataset_train:
|
|
batch = static_dataset_train.expand_batch(batch_to_dict(batch, signature_dynamic))
|
|
uv_curvature = tf.one_hot(batch['uv_curvature'], 7)
|
|
nr_of_samples = 16
|
|
|
|
# pred = model.sample(batch['uv_static'], batch['uv_body_posed'], batch['uv_mask'])
|
|
pred = model.sample_single(batch['uv_static'][0], batch['uv_body_posed'][0], batch['uv_mask'][0], nr_of_samples=nr_of_samples)
|
|
|
|
ce_uv_loss = scce(tf.tile(batch['uv_curvature'][0][tf.newaxis,:,:], [nr_of_samples,1,1]), pred)
|
|
|
|
|
|
mesh_pred = uv2mesh(pred).numpy().argmax(axis=-1)
|
|
mesh_orig = batch['mesh_curvature'][0].numpy()
|
|
|
|
F = np.load("custom_utils/smpl_faces.npy")
|
|
mask = batch['mesh_mask'].numpy().astype(bool)[0]
|
|
_, faces = _maskFaces(np.ones((14475, 3)), F, mask)
|
|
mesh_pred = [mesh_pred[i, mask] for i in range(mesh_pred.shape[0])]
|
|
curv_color = [np.zeros((mesh_pred[i].shape[0], 3)) for i in range(len(mesh_pred))]
|
|
mesh_orig = mesh_orig[mask] + 1
|
|
|
|
colors = np.array([cm.color('jet', 1 - i) for i in np.linspace(0, 1, 6+1)])
|
|
curv_color_gt = np.zeros((mesh_pred[0].shape[0], 3))
|
|
for i in range(6+1):
|
|
for j in range(len(mesh_pred)):
|
|
curv_color[j][mesh_pred[j] == i, :] = colors[i]
|
|
curv_color_gt[mesh_orig == i, :] = colors[i]
|
|
|
|
mesh_base = batch['mesh_posed'][0].numpy()[mask, :]
|
|
meshes = []
|
|
meshes.append(go.Mesh3d(
|
|
# 8 vertices of a cube
|
|
x=mesh_base[:,0],
|
|
y=mesh_base[:,1],
|
|
z=mesh_base[:,2],
|
|
# i, j and k give the vertices of triangles
|
|
i = faces[:,0],
|
|
j = faces[:,1],
|
|
k = faces[:,2],
|
|
vertexcolor=curv_color_gt,
|
|
text='gt'
|
|
))
|
|
|
|
for i in range(nr_of_samples):
|
|
x = mesh_base.copy()
|
|
x[:, 1] += 1 * (i + 1)
|
|
meshes.append(go.Mesh3d(
|
|
# 8 vertices of a cube
|
|
x=x[:,0],
|
|
y=x[:,1],
|
|
z=x[:,2],
|
|
# i, j and k give the vertices of triangles
|
|
i = faces[:,0],
|
|
j = faces[:,1],
|
|
k = faces[:,2],
|
|
vertexcolor=curv_color[i]
|
|
))
|
|
|
|
|
|
|
|
|
|
fig = go.Figure(data=meshes)
|
|
|
|
fig.update_layout(scene_aspectmode='manual', scene_aspectratio=dict(x=1, y=nr_of_samples-1, z=1))
|
|
fig.update_layout(
|
|
scene = dict(
|
|
xaxis = dict(nticks=4, range=[-1,1],),
|
|
yaxis = dict(nticks=4, range=[-1,nr_of_samples + 1],),
|
|
zaxis = dict(nticks=4, range=[-1,1],),)
|
|
)
|
|
|
|
# fig.show()
|
|
|
|
pred = pred.numpy().argmax(-1)
|
|
frames = [Image.fromarray(pred[i]/6*255) for i in range(pred.shape[0])]
|
|
frames[0].save("temp/curv2.gif", format="GIF", append_images=frames[1:],
|
|
save_all=True, duration=500, loop=0)
|
|
|
|
print()
|
|
|
|
if __name__ == '__main__':
|
|
init_wandb()
|
|
allow_memory_growth()
|
|
parser = argparse.ArgumentParser(description='')
|
|
parser.add_argument('--name', type=str)
|
|
args = vars(parser.parse_args())
|
|
|
|
batch_size = 4
|
|
|
|
smpl = SMPL()
|
|
scce = tf.keras.losses.SparseCategoricalCrossentropy()
|
|
|
|
factory = DatasetFactory()
|
|
signature_dynamic = { 'id': (),
|
|
'pose_shape': (72 + 11,),
|
|
# 'uv_posed': (3, 512, 256,),
|
|
'uv_body_posed': (3, 512, 256,),
|
|
'mesh_body_posed': (14475, 3,),
|
|
'mesh_posed': (14475, 3,),
|
|
# 'mesh_unposed': (14475, 3,),
|
|
# 'uv_unposed': (3, 512, 256,),
|
|
'uv_curvature': (512, 256),
|
|
'mesh_curvature': (14475)
|
|
}
|
|
signature_static = { 'uv_static': (3, 512, 256,),
|
|
# 'mesh_static': (14475, 3,),
|
|
'mesh_mask': (14475,),
|
|
'uv_mask': (512, 256,),
|
|
# 'uv_shaped_body': (3, 512, 256,),
|
|
}
|
|
|
|
dataset_train = factory.get_dataset(folder='/home/workspace/local-data/train_100', batch_size=batch_size, signature=signature_dynamic, use_filled=False, resolutions=[512])
|
|
static_dataset_train = StaticDataset(folder='/home/workspace/local-data/train_100', signature=signature_static, use_filled=False, resolutions=[512])
|
|
dataset_valid = factory.get_dataset(folder='/home/workspace/local-data/jumpsuit_20/valid', batch_size=batch_size, signature=signature_dynamic)
|
|
static_dataset_valid = StaticDataset(folder='/home/workspace/local-data/jumpsuit_20/valid', signature=signature_static)
|
|
|
|
|
|
model = CVAE(latent_dim=128, static_model='checkpoints/ae_static_5',
|
|
body_model='checpoints/ae_body_8',)
|
|
|
|
|
|
# lr_schedule = tf.keras.optimizers.schedules.ExponentialDecay(0.0001, decay_steps=30000, decay_rate=0.1, staircase=False)
|
|
optimizer = tf.optimizers.Adam(learning_rate=0.0001)
|
|
balancing_coeff = 0.1
|
|
reg_coeff = 1
|
|
folder = 'models/ae_logs/'
|
|
valid_iterator = iter(dataset_valid)
|
|
gamma = 1
|
|
balancing_activated_step = 0
|
|
|
|
|
|
# sample('checkpoints/cvae_curv_2')
|
|
|
|
|
|
ckpt = tf.train.Checkpoint(model=model)
|
|
manager = tf.train.CheckpointManager(ckpt, 'checkpoints/' + name, max_to_keep=3)
|
|
|
|
step = 0
|
|
for batch in dataset_train:
|
|
batch = static_dataset_train.expand_batch(batch_to_dict(batch, signature_dynamic))
|
|
|
|
train_loss = train_step(batch)
|
|
|
|
# data_valid = valid_iterator.get_next()
|
|
# data_valid = static_dataset_valid.expand_batch(batch_to_dict(data_valid, signature_dynamic))
|
|
# valid_loss = valid_step(data_valid)
|
|
valid_loss = {}
|
|
|
|
to_log = { **train_loss, **valid_loss, **{ 'current_step': step} }
|
|
if step % 2500 == 0:
|
|
train_figs = plot_everything(batch, 'train')
|
|
# valid_figs = plot_everything(data_valid, 'valid')
|
|
valid_figs = {}
|
|
to_log = { **to_log, **train_figs, **valid_figs }
|
|
|
|
if step % 25000 == 0:
|
|
manager.save(checkpoint_number=step)
|
|
|
|
if step > 25000:
|
|
# if train_loss['train/euclid_error'] < 0.01:
|
|
balancing_coeff = balancing_coeff + 1 / 20000
|
|
# if train_loss['train/euclid_error'] < 0.015:
|
|
# reg_coeff = reg_coeff + 100 / 15000
|
|
|
|
|
|
# if step > 15000 and balancing_coeff <= 1e-3:
|
|
# balancing_coeff = 1e-2
|
|
# if step > 20000 and balancing_coeff <= 1e-3:
|
|
# balancing_coeff = 1e-2
|
|
# if step > 25000 and balancing_coeff <= 1e-2:
|
|
# balancing_coeff = 1e-1
|
|
# if step > 30000 and balancing_coeff <= 1e-1:
|
|
# balancing_coeff = 1
|
|
|
|
# gamma = 0.99 * gamma + 0.01 * train_loss['train/reconstr_loss']
|
|
|
|
to_log = { **to_log, **{ 'train/balancing_coeff': balancing_coeff, 'train/gamma': gamma } }
|
|
|
|
wandb.log(to_log, step = step)
|
|
|
|
print(step)
|
|
|
|
step += 1 |