forked from daviddao/spatial-transformer-tensorflow
-
Notifications
You must be signed in to change notification settings - Fork 5
/
test_elastic.py
72 lines (62 loc) · 2.71 KB
/
test_elastic.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
# Copyright 2016 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
from scipy import ndimage
import tensorflow as tf
from spatial_transformer import ElasticTransformer
import numpy as np
import scipy.misc
# Input image retrieved from:
# https://raw.githubusercontent.com/skaae/transformer_network/master/cat.jpg
#im = ndimage.imread('data/cat.jpg')
im = ndimage.imread('data/cat_small.png')
im = im / 255.
im = im.astype('float32')
# input batch
batch_size = 4
batch = np.expand_dims(im, axis=0)
batch = np.tile(batch, [batch_size, 1, 1, 1])
# input placeholder
x = tf.placeholder(tf.float32, [batch_size, im.shape[0], im.shape[1], im.shape[2]])
# Let the output size of the projective transformer be quarter of the image size.
outsize = (int(im.shape[0]), int(im.shape[1]))
# Elastic Transformation Layer
stl = ElasticTransformer(outsize)
theta = tf.get_variable('theta', [batch_size, stl.param_dim], trainable=False)
# Run session
with tf.Session() as sess:
with tf.device("/cpu:0"):
with tf.variable_scope('spatial_transformer'):
# Random jitter of identity parameters
cur_theta = 0.1*np.random.randn(batch_size, stl.param_dim)
#cur_theta = 0.0 + 0.2*np.random.randn(1, stl.param_dim)
#cur_theta = np.repeat(cur_theta, batch_size, axis=0)
side_dim = np.floor(np.sqrt(stl.param_dim//2))
for i in range(stl.param_dim//2):
if i<side_dim or i>=(side_dim-1)*side_dim or i%side_dim==0 or i%side_dim==side_dim-1:
#print(i)
cur_theta[:,i] = 0.0
cur_theta[:,stl.param_dim//2 + i] = 0.0
result = stl.transform(x, theta)
inv_result = stl.transform(x, theta, False)
# %% Run session
sess.run(tf.global_variables_initializer())
result_ = sess.run(result, feed_dict={x: batch, theta: cur_theta})
result__ = sess.run(inv_result, feed_dict={x: result_, theta: cur_theta})
# save our result_
for i in range(result_.shape[0]):
scipy.misc.imsave('elastic' + str(i) + '.png', result_[i])
# save our result__
for i in range(result__.shape[0]):
scipy.misc.imsave('elasticr' + str(i) + '.png', result__[i])