def main():
from qpwcnet.core.util import disable_gpu
from qpwcnet.data.triplet_dataset_ops import show_triplet_dataset
disable_gpu()
dataset = VimeoTriplet(VimeoTripletSettings(data_type='train'))
show_triplet_dataset(dataset)
def main():
disable_gpu()
# compute_stats()
if False:
filename = '/media/ssd/datasets/sintel-processed/sintel.tfrecord'
reader = get_reader(filename).map(preprocess)
else:
reader = get_dataset_from_set().map(preprocess_fc3d)
# reader = get_dataset().interleave(lambda x: Dataset.from_tensors(x).map(decode_files),
# cycle_length=tf.data.experimental.AUTOTUNE,
# num_parallel_calls=tf.data.experimental.AUTOTUNE).map(preprocess)
reader.shuffle(buffer_size=32)
for entry in reader.as_numpy_iterator():
ims, flo = entry
flo_vis = flow_to_image(flo)
prv = ims[..., :3]
nxt = ims[..., 3:]
#print('prv', prv.min(), prv.max())
#print('nxt', nxt.min(), nxt.max())
#print('flo', flo.min(), flo.max())
#print('flo', np.linalg.norm(flo, axis=-1).mean())
# show prev reconstructed from nxt.
# nxt_w = tfa.image.dense_image_warp(nxt[None, ...].astype(
# np.float32)/255.0, -flo[None, ..., ::-1]).numpy()
# nxt_w = tf_warp(nxt[None, ...].astype(
# np.float32)/255.0, flo[None, ...]).numpy()
# flo order : (x,y) == (1,0)
nxt_w = tfa.image.dense_image_warp(nxt[None, ...],
-flo[None, ..., ::-1])[0].numpy()
# nxt_w = tf_warp(nxt[None, ...], flo)[0].numpy()
print(nxt_w.shape)
cv2.imshow('prv', prv)
cv2.imshow('nxt', nxt)
# cv2.imshow('msk', prv_has_flo.astype(np.float32))
cv2.imshow('nxt_w', nxt_w)
cv2.imshow('nxt_w2', nxt_w - prv)
# bgr, prv=b, nxt=g, r=warp
overlay = np.stack([(prv).mean(axis=-1), (nxt).mean(axis=-1),
(nxt_w).mean(axis=-1)],
axis=-1)
cv2.imshow('overlay', overlay)
cv2.imshow('flo', normalize(flo[..., 0]))
cv2.imshow('flo-vis', flo_vis.numpy())
k = cv2.waitKey(0)
if k == 27:
break
def main():
disable_gpu()
data_format = 'channels_first'
tf.keras.backend.set_image_data_format(data_format)
dataset = setup_input(8, data_format)
for imgs, flows in dataset:
idx = np.random.randint(8)
prv = imgs[idx, :3]
nxt = imgs[idx, 3:]
flo = flows[idx]
flo_rgb = flow_to_image(flo, data_format=data_format)
show('prv', 0.5 + prv, True, data_format)
show('nxt', 0.5 + nxt, True, data_format)
# FLO corresponds to stuff in `prv`
show('flo', flo_rgb, True, data_format)
k = cv2.waitKey(0)
if k in [27, ord('q')]:
break
def main():
disable_gpu()
data_format = 'channels_last'
warp_1 = Warp(data_format=data_format)
warp_2 = WarpV2(data_format=data_format)
if data_format == 'channels_first':
img = tf.random.uniform(shape=(4, 3, 32, 64))
flo = tf.random.normal(shape=(4, 2, 32, 64))
c1 = warp_1((img, flo))
c2 = warp_2((img, flo))
print('diff', tf.reduce_mean(c1 - c2))
else:
img = tf.random.uniform(shape=(4, 32, 64, 3))
flo = tf.random.normal(shape=(4, 32, 64, 2))
c1 = warp_1((img, flo))
c2 = warp_2((img, flo))
print('diff', tf.reduce_mean(c1 - c2))
cv2.imshow('c1', c1[0].numpy())
cv2.imshow('c2', c2[0].numpy())
cv2.imshow('diff', tf.abs(c1 - c2)[0].numpy())
cv2.waitKey(0)
ims, flo = image_resize(ims, flo, (256, 512))
ims = ims - 0.5
# Convert to correct data format
data_format = tf.keras.backend.image_data_format()
if data_format == 'channels_first':
ims = einops.rearrange(ims, '... h w c -> ... c h w')
flo = einops.rearrange(flo, '... h w c -> ... c h w')
return ims, flo
if True:
tf.keras.backend.set_image_data_format('channels_last')
data_format = tf.keras.backend.image_data_format()
disable_gpu()
# TODO(ycho): Cleanup dataset loading pattern for opt-flow datasets.
glob_pattern = '/media/ssd/datasets/sintel-processed/shards/sintel-*.tfrecord'
filenames = tf.data.Dataset.list_files(glob_pattern).shuffle(32)
# dataset = get_reader(filenames).shuffle(buffer_size=1024).repeat().batch(8)
# dataset = get_reader(filenames).batch(8).repeat()
dataset = get_reader(filenames).shuffle(
buffer_size=32).map(preprocess).batch(1)
for ims, flo in dataset:
# Unstack `ims`.
if data_format == 'channels_first':
prv, nxt = einops.rearrange(
ims, 'n (k c) h w -> k n c h w', k=2)
else: