if len(sys.argv) > 1:
logdir = sys.argv[1]
mkdir(logdir)
sys.stdout = open(str(logdir) + '/log.txt', 'w')
"""Architecture configuration."""
num_blocks = 8
batch_size = 24
input_shape = mpii_sp_dataconf.input_shape
num_joints = 16
mode = VALID_MODE
# Build the model
model = reception.build(input_shape,
num_joints,
dim=2,
num_blocks=num_blocks,
num_context_per_joint=2,
ksize=(5, 5),
concat_pose_confidence=False)
weights_path = get_file(weights_file,
TF_WEIGHTS_PATH,
file_hash=md5_hash,
cache_subdir='models')
"""Load pre-trained model."""
model.load_weights(weights_path)
"""Merge pose and visibility as a single output."""
outputs = []
for b in range(int(len(model.outputs) / 2)):
outputs.append(
concatenate([model.outputs[2 * b], model.outputs[2 * b + 1]],
mkdir(logdir)
sys.stdout = open(str(logdir) + '/log.txt', 'w')
num_frames = 20
num_blocks = 4
batch_size = 2
depth_maps = 8
num_joints = 20
num_actions = 60
pose_dim = 3
input_shape = ntu_dataconf.input_shape
"""Build the pose estimation model."""
model_pe = reception.build(input_shape,
num_joints,
dim=pose_dim,
num_blocks=num_blocks,
depth_maps=depth_maps,
ksize=(5, 5),
concat_pose_confidence=False)
"""Build the full model using the previous pose estimation one."""
model = action.build_merge_model(model_pe,
num_actions,
input_shape,
num_frames,
num_joints,
num_blocks,
pose_dim=pose_dim,
num_context_per_joint=0,
pose_net_version='v2')
"""Load pre-trained model."""
weights_path = get_file(weights_file,
logdir = sys.argv[1]
mkdir(logdir)
sys.stdout = open(str(logdir) + '/log.txt', 'w')
print("*******debug******")
weights_file = os.path.join(logdir, 'weights_mpii_{epoch:03d}.h5')
"""Architecture configuration."""
num_blocks = 8
batch_size = 24
input_shape = mpii_sp_dataconf.input_shape
num_joints = 16
print("***********************")
print(input_shape)
model = reception.build(input_shape,
num_joints,
dim=2,
num_blocks=num_blocks,
num_context_per_joint=2,
ksize=(5, 5))
"""Load the MPII dataset."""
mpii = MpiiSinglePerson('datasets/MPII', dataconf=mpii_sp_dataconf)
data_tr = BatchLoader(mpii, ['frame'], ['pose'],
TRAIN_MODE,
batch_size=batch_size,
num_predictions=num_blocks,
shuffle=True)
"""Pre-load validation samples and generate the eval. callback."""
mpii_val = BatchLoader(mpii,
x_dictkeys=['frame'],
y_dictkeys=['pose', 'afmat', 'headsize'],
mode=VALID_MODE,
logdir = sys.argv[1]
mkdir(logdir)
sys.stdout = open(str(logdir) + '/log.txt', 'w')
num_frames = 20
num_blocks = 4
batch_size = 2
depth_maps = 8
num_joints = 20
num_actions = 60
pose_dim = 3
input_shape = ntu_ar_dataconf.input_shape
"""Build the pose estimation model."""
model_pe = reception.build(input_shape,
num_joints,
dim=pose_dim,
num_blocks=num_blocks,
depth_maps=depth_maps,
ksize=(5, 5))
"""Build the full model using the previous pose estimation one."""
model = action.build_merge_model(model_pe,
num_actions,
input_shape,
num_frames,
num_joints,
num_blocks,
pose_dim=pose_dim,
num_context_per_joint=0,
pose_net_version='v2')
"""Load pre-trained model."""
model.load_weights('weights_0052.h5')
"""Load NTU dataset."""
md5_hash = 'af79f83ad939117d4ccc2cf1d4bd37d2'
logdir = './'
if len(sys.argv) > 1:
logdir = sys.argv[1]
mkdir(logdir)
sys.stdout = open(str(logdir) + '/log.txt', 'w')
num_blocks = 8
batch_size = 24
input_shape = human36m_dataconf.input_shape
num_joints = pa17j3d.num_joints
model = reception.build(input_shape,
num_joints,
dim=3,
num_blocks=num_blocks,
ksize=(5, 5))
"""Load pre-trained model."""
weights_path = get_file(weights_file,
TF_WEIGHTS_PATH,
md5_hash=md5_hash,
cache_subdir='models')
model.load_weights(weights_path)
"""Merge pose and visibility as a single output."""
outputs = []
for b in range(int(len(model.outputs) / 2)):
outputs.append(
concatenate([model.outputs[2 * b], model.outputs[2 * b + 1]],
name='blk%d' % (b + 1)))
model = Model(model.input, outputs, name=model.name)
