-
Notifications
You must be signed in to change notification settings - Fork 6
/
ops.py
86 lines (61 loc) · 3.29 KB
/
ops.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
import math
import numpy as np
import tensorflow as tf
import tensorflow.contrib.slim as slim
from tensorflow.python.framework import ops
from utils import *
def batch_norm(x, name="batch_norm"):
return tf.contrib.layers.batch_norm(x, decay=0.9, updates_collections=None, epsilon=1e-5, scale=True, scope=name)
def instance_norm(input, name="instance_norm"):
with tf.variable_scope(name):
depth = input.get_shape()[3]
scale = tf.get_variable("scale", [depth], initializer=tf.random_normal_initializer(1.0, 0.02, dtype=tf.float32))
offset = tf.get_variable("offset", [depth], initializer=tf.constant_initializer(0.0))
mean, variance = tf.nn.moments(input, axes=[1,2], keep_dims=True)
epsilon = 1e-5
inv = tf.rsqrt(variance + epsilon)
normalized = (input-mean)*inv
return scale*normalized + offset
def conv2d(input_, output_dim, ks=4, s=2, stddev=0.02, padding='SAME', name="conv2d"):
with tf.variable_scope(name):
return slim.conv2d(input_, output_dim, ks, s, padding=padding, activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=stddev),
biases_initializer=None)
def deconv2d(input_, output_dim, ks=4, s=2, stddev=0.02, name="deconv2d"):
with tf.variable_scope(name):
return slim.conv2d_transpose(input_, output_dim, ks, s, padding='SAME', activation_fn=None,
weights_initializer=tf.truncated_normal_initializer(stddev=stddev),
biases_initializer=None)
def lrelu(x, leak=0.2, name="lrelu"):
return tf.maximum(x, leak*x)
def linear(input_, output_size, scope=None, stddev=0.02, bias_start=0.0, with_w=False):
with tf.variable_scope(scope or "Linear"):
matrix = tf.get_variable("Matrix", [input_.get_shape()[-1], output_size], tf.float32,
tf.random_normal_initializer(stddev=stddev))
bias = tf.get_variable("bias", [output_size],
initializer=tf.constant_initializer(bias_start))
if with_w:
return tf.matmul(input_, matrix) + bias, matrix, bias
else:
return tf.matmul(input_, matrix) + bias
def tf_image_translate(input_, tx, ty, interpolation='NEAREST'):
# got these parameters from solving the equations for pixel translations
# on https://www.tensorflow.org/api_docs/python/tf/contrib/image/transform
transforms = [1, 0, -tx, 0, 1, -ty, 0, 0]
return tf.contrib.image.transform(input_, transforms, interpolation)
def gaussian_kernel(sigma, sz):
xpos_vec = np.arange(sz)
ypos_vec = np.arange(sz)
output = np.ones([sz, sz, 1, 1], dtype=np.single)
midpos = sz // 2
for xpos in xpos_vec:
for ypos in ypos_vec:
output[xpos,ypos,:,:] = np.exp(-((xpos-midpos)**2 + (ypos-midpos)**2) / (2 * sigma**2)) / (2 * np.pi * sigma**2)
return output
def Dp(image, xshift, yshift, option):
shift_image = tf_image_translate(image, xshift, yshift, interpolation='NEAREST')
diff = tf.subtract(image, shift_image)
diff_square = tf.multiply(diff, diff)
res = tf.nn.conv2d(diff_square, filter=gaussian_kernel(option.sigma, option.patch_size), strides=[1,1,1,1],
padding="SAME", name="patch_distance")
return res