-
Notifications
You must be signed in to change notification settings - Fork 22
/
loss_utils.py
207 lines (158 loc) · 6.43 KB
/
loss_utils.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
import tensorflow as tf
import numpy as np
import tensorflow.contrib.slim as slim
import cv2
# Some were adopted from
# https://github.com/tensorflow/models/tree/master/research/video_prediction
def average_gradients(tower_grads):
"""Calculate the average gradient for each shared variable across all towers.
Note that this function provides a synchronization point across all towers.
Args:
tower_grads: List of lists of (gradient, variable) tuples. The outer list
is over individual gradients. The inner list is over the gradient
calculation for each tower.
Returns:
List of pairs of (gradient, variable) where the gradient has been averaged
across all towers.
"""
average_grads = []
for grad_and_vars in zip(*tower_grads):
# Note that each grad_and_vars looks like the following:
# ((grad0_gpu0, var0_gpu0), ... , (grad0_gpuN, var0_gpuN))
grads = []
for g, _ in grad_and_vars:
# Add 0 dimension to the gradients to represent the tower.
expanded_g = tf.expand_dims(g, 0)
# Append on a 'tower' dimension which we will average over below.
grads.append(expanded_g)
# Average over the 'tower' dimension.
grad = tf.concat(axis=0, values=grads)
grad = tf.reduce_mean(grad, 0)
# Keep in mind that the Variables are redundant because they are shared
# across towers. So .. we will just return the first tower's pointer to
# the Variable.
v = grad_and_vars[0][1]
grad_and_var = (grad, v)
average_grads.append(grad_and_var)
return average_grads
def mean_squared_error(true, pred):
"""L2 distance between tensors true and pred.
Args:
true: the ground truth image.
pred: the predicted image.
Returns:
mean squared error between ground truth and predicted image.
"""
return tf.reduce_sum(tf.square(true - pred)) / tf.to_float(tf.size(pred))
def weighted_mean_squared_error(true, pred, weight):
"""L2 distance between tensors true and pred.
Args:
true: the ground truth image.
pred: the predicted image.
Returns:
mean squared error between ground truth and predicted image.
"""
tmp = tf.reduce_sum(
weight * tf.square(true - pred), axis=[1, 2],
keep_dims=True) / tf.reduce_sum(
weight, axis=[1, 2], keep_dims=True)
return tf.reduce_mean(tmp)
def mean_L1_error(true, pred):
"""L2 distance between tensors true and pred.
Args:
true: the ground truth image.
pred: the predicted image.
Returns:
mean squared error between ground truth and predicted image.
"""
return tf.reduce_sum(tf.abs(true - pred)) / tf.to_float(tf.size(pred))
def weighted_mean_L1_error(true, pred, weight):
"""L2 distance between tensors true and pred.
Args:
true: the ground truth image.
pred: the predicted image.
Returns:
mean squared error between ground truth and predicted image.
"""
return tf.reduce_sum(tf.abs(true - pred) *
weight) / tf.to_float(tf.size(pred))
def cal_grad2_error(flo, image, beta):
"""
Calculate the image-edge-aware second-order smoothness loss for flo
"""
def gradient(pred):
D_dy = pred[:, 1:, :, :] - pred[:, :-1, :, :]
D_dx = pred[:, :, 1:, :] - pred[:, :, :-1, :]
return D_dx, D_dy
img_grad_x, img_grad_y = gradient(image)
weights_x = tf.exp(-10.0 * tf.reduce_mean(
tf.abs(img_grad_x), 3, keep_dims=True))
weights_y = tf.exp(-10.0 * tf.reduce_mean(
tf.abs(img_grad_y), 3, keep_dims=True))
dx, dy = gradient(flo)
dx2, dxdy = gradient(dx)
dydx, dy2 = gradient(dy)
return (tf.reduce_mean(beta*weights_x[:,:, 1:, :]*tf.abs(dx2)) + \
tf.reduce_mean(beta*weights_y[:, 1:, :, :]*tf.abs(dy2))) / 2.0
def cal_grad2_error_mask(flo, image, beta, mask):
"""
Calculate the image-edge-aware second-order smoothness loss for flo
within the given mask
"""
def gradient(pred):
D_dy = pred[:, 1:, :, :] - pred[:, :-1, :, :]
D_dx = pred[:, :, 1:, :] - pred[:, :, :-1, :]
return D_dx, D_dy
img_grad_x, img_grad_y = gradient(image)
weights_x = tf.exp(-10.0 * tf.reduce_mean(
tf.abs(img_grad_x), 3, keep_dims=True))
weights_y = tf.exp(-10.0 * tf.reduce_mean(
tf.abs(img_grad_y), 3, keep_dims=True))
dx, dy = gradient(flo)
dx2, dxdy = gradient(dx)
dydx, dy2 = gradient(dy)
return (tf.reduce_mean(beta*weights_x[:,:, 1:, :]*tf.abs(dx2) * mask[:, :, 1:-1, :]) + \
tf.reduce_mean(beta*weights_y[:, 1:, :, :]*tf.abs(dy2) * mask[:, 1:-1, :, :])) / 2.0
def SSIM(x, y):
C1 = 0.01**2
C2 = 0.03**2
mu_x = slim.avg_pool2d(x, 3, 1, 'VALID')
mu_y = slim.avg_pool2d(y, 3, 1, 'VALID')
sigma_x = slim.avg_pool2d(x**2, 3, 1, 'VALID') - mu_x**2
sigma_y = slim.avg_pool2d(y**2, 3, 1, 'VALID') - mu_y**2
sigma_xy = slim.avg_pool2d(x * y, 3, 1, 'VALID') - mu_x * mu_y
SSIM_n = (2 * mu_x * mu_y + C1) * (2 * sigma_xy + C2)
SSIM_d = (mu_x**2 + mu_y**2 + C1) * (sigma_x + sigma_y + C2)
SSIM = SSIM_n / SSIM_d
return tf.clip_by_value((1 - SSIM) / 2, 0, 1)
def deprocess_image(image):
# Assuming input image is float32
return tf.image.convert_image_dtype(image, dtype=tf.uint8)
def preprocess_image(image):
# Assuming input image is uint8
image = tf.image.convert_image_dtype(image, dtype=tf.float32)
return image
def charbonnier_loss(x,
mask=None,
truncate=None,
alpha=0.45,
beta=1.0,
epsilon=0.001):
"""Compute the generalized charbonnier loss of the difference tensor x.
All positions where mask == 0 are not taken into account.
Args:
x: a tensor of shape [num_batch, height, width, channels].
mask: a mask of shape [num_batch, height, width, mask_channels],
where mask channels must be either 1 or the same number as
the number of channels of x. Entries should be 0 or 1.
Returns:
loss as tf.float32
"""
batch, height, width, channels = tf.unstack(tf.shape(x))
normalization = tf.cast(batch * height * width * channels, tf.float32)
error = tf.pow(tf.square(x * beta) + tf.square(epsilon), alpha)
if mask is not None:
error = tf.multiply(mask, error)
if truncate is not None:
error = tf.minimum(error, truncate)
return tf.reduce_sum(error) / normalization