-
Notifications
You must be signed in to change notification settings - Fork 2
/
deep_resnet_subpixel.py
222 lines (175 loc) · 6.27 KB
/
deep_resnet_subpixel.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
208
209
210
211
212
213
214
# -*- coding: utf-8 -*-
from keras.layers import *
from keras.models import *
from keras.optimizers import *
from keras.callbacks import *
import keras.backend as K
import tensorflow as tf
import matplotlib.pyplot as plt
from data_gen import DataGen
import numpy as np
from sklearn.model_selection import train_test_split
#db_path = r'G:\mix5k.h5'
db_path = r'E:\images_uint8.h5'
train_split = .75
ishape = (16,16)
pshape = (64,64)
strides = (32,32)
bsize = 64
num_epochs=30
dg = DataGen(db_path, pshape=pshape, strides=strides, ds_method='cubic')
with dg.load_db() as db:
all_data=dg.get_datasets()
train,test=train_test_split(all_data, train_size=train_split,
shuffle=True, random_state=7)
train_patches = dg.get_patch_list(train)
test_patches = dg.get_patch_list(test, pshape=pshape, strides=strides)
print('Loading Images ....')
dg.load_images()
train_gen = dg.patch_gen(train_patches, bsize)
test_gen = dg.patch_gen(test_patches, bsize)
val_gen = dg.patch_gen(test_patches, bsize)
len_train = len(train_patches)//bsize
len_test = len(test_patches)//bsize
len_train = len_train//5
len_test = len_test//5
def prop_sigmoid(x):
over = tf.stop_gradient(tf.maximum(x-1., 0.))
under = tf.stop_gradient(tf.maximum(0.-x, 0.))
return x-over+under
def subpixelate(x):
f = 4
b,h,w,c=tf.unstack(tf.shape(x))
tc = x.get_shape().as_list()[-1]//(f*f)
x = tf.reshape(x, [b,h,w,f,f,tc])
x = tf.transpose(x, [0,5,1,3,2,4])
x = tf.reshape(x, [b,tc,h*f,w*f])
x = tf.transpose(x, [0,2,3,1])
return x
K.clear_session()
in_t = Input(shape=(None,None,3), dtype='int8')
in_tx1 = Lambda(lambda v: tf.cast(v,tf.float32)/255.)(in_t)
x = in_tx1
x = Conv2D(128, (5,5), padding='same', activation=None)(x)
x = Activation('relu')(x)
y = x
x = Concatenate()([x,in_tx1])
x = Conv2D(128, (3,3), padding='same', activation=None)(x)
x = Add()([x,y])
x = Activation('relu')(x)
y = x
x = Concatenate()([x,in_tx1])
x = Conv2D(128, (3,3), padding='same', dilation_rate=1,activation=None)(x)
x = Add()([x,y])
x = Activation('relu')(x)
y = x
y = Conv2D(256, (1,1), activation=None)(y)
x = Concatenate()([x,in_tx1])
x = Conv2D(256, (3,3), padding='same', dilation_rate=1, activation=None)(x)
x = Add()([x,y])
x = Activation('relu')(x)
y = x
x = Concatenate()([x,in_tx1])
x = Conv2D(256, (3,3), padding='same', dilation_rate=1, activation=None)(x)
x = Add()([x,y])
x = Activation('relu')(x)
y = x
y = Conv2D(512, (1,1), activation=None)(y)
x = Concatenate()([x,in_tx1])
x = Conv2D(512, (3,3), padding='same', dilation_rate=2, activation=None)(x)
x = Add()([x,y])
x = Activation('relu')(x)
y = x
x = Concatenate()([x,in_tx1])
x = Conv2D(512, (3,3), padding='same', dilation_rate=2, activation=None)(x)
x = Add()([x,y])
x = Activation('relu')(x)
x = Conv2D(3*16, (3,3), padding='same', dilation_rate=2, activation=None)(x)
x = Lambda(subpixelate)(x)
x = Activation(prop_sigmoid)(x)
out_t = x
def int_mse_loss(y_true, y_pred):
print(y_true.dtype)
return tf.reduce_mean((y_true/255.-y_pred)**2)
model = Model(in_t, out_t)
print(model.summary())
opt = Adam(lr=1e-4)
model.compile(opt, int_mse_loss)
def on_ep_end(epoch, log):
plt.figure(figsize=(15,5))
X, Y = next(val_gen)
Y_pred = model.predict_on_batch(X)
for k in range(15):
c, r = k//5, k%5
dp = r*3+c
sp = r*11+c*4
plt.subplot(5, 11, 1 + sp)
plt.axis('off')
plt.imshow(X[dp])
if not r: plt.title('LR')
plt.subplot(5, 11, 2 + sp)
plt.axis('off')
plt.imshow(np.clip(Y_pred[dp],0.,1.))
if not r: plt.title('HRPred')
plt.subplot(5, 11, 3 + sp)
plt.axis('off')
plt.imshow(Y[dp])
if not r: plt.title('HR')
plt.pause(.01)
figs=plt.get_fignums()
if len(figs)>15:
for ff in figs[0:len(figs)-15]: plt.close(ff)
ep_end = LambdaCallback(on_epoch_end=on_ep_end)
mchk = ModelCheckpoint('deep_resnet_subpix_checkpoint.h5',save_best_only=True)
redlr = ReduceLROnPlateau(factor=.75, patience=1, verbose=1, cooldown=1)
#model.load_weights('big_subpix_deep_resnet_incremental_v1_32x32_loss0038_linear_downsampling_imagenet_validationdata.h5')
model.fit_generator(train_gen, steps_per_epoch=len_train, epochs=num_epochs,
validation_data=test_gen, validation_steps=len_test/4,
callbacks=[ep_end, mchk])
model.save('big_subpix_deep_resnet_incremental_v1_32x32_loss0038_linear_downsampling_imagenet_validationdata.h5')
import cv2
img=dg.image_dict[test[5]]
imgl=cv2.resize(img, None, fx=.25, fy=.25)
imgs=model.predict_on_batch(imgl[None,...])[0]
imgr=resize(img, imgs.shape).astype('float32')
plt.figure(figsize=[7,7]), plt.axis('off')
plt.imshow(imgl)
plt.title('LO-RES')
plt.tight_layout()
plt.figure(figsize=[20,7])
plt.subplot(131),plt.axis('off')
imgc=np.clip(cv2.resize(imgl/255., None, fx=4, fy=4, interpolation=cv2.INTER_CUBIC), 0.,1.)
plt.imshow(imgc)
plt.title('4xBI-CUBIC')
plt.subplot(132),plt.axis('off')
plt.imshow(imgs)
plt.title('4xSUPERRESOLUTION')
plt.subplot(133),plt.axis('off')
plt.imshow(imgr)
plt.title('ORIGINAL HIRES')
plt.tight_layout()
num_img = len(test)
bic_psnr_arr, bic_ssim_arr = [], [];
sup_psnr_arr, sup_ssim_arr = [], [];
for k in range(num_img):
if not k%10:
print('{}/{}'.format(k,num_img))
img=dg.image_dict[test[k]]
if img.shape[0]*img.shape[1]>400*500:
continue
imgl=cv2.resize(img, None, fx=.25, fy=.25)
imgs=model.predict_on_batch(imgl[None,...])[0]
imgr=resize(img, imgs.shape).astype('float32')
imgc=np.clip(cv2.resize(imgl/255., None, fx=4, fy=4, interpolation=cv2.INTER_CUBIC), 0.,1.).astype('float32')
bic_psnr_arr.append(compare_psnr(imgr, imgc))
sup_psnr_arr.append(compare_psnr(imgr, imgs))
bic_ssim_arr.append(compare_ssim(imgr, imgc, multichannel=True, data_range=1.0, gaussian_weights=True))
sup_ssim_arr.append(compare_ssim(imgr, imgs, multichannel=True, data_range=1.0, gaussian_weights=True))
sup_avg_psnr=np.mean(sup_psnr_arr)
sup_avg_ssim = np.mean(sup_ssim_arr)
bic_avg_psnr = np.mean(bic_psnr_arr)
bic_avg_ssim = np.mean(bic_ssim_arr)
print('Bi-cubic PSNR:', bic_avg_psnr)
print('Superresolution PSNR:', sup_avg_psnr)
print('Bi-cubic SSIM:', bic_avg_ssim)
print('Superresolution SSIM:', sup_avg_ssim)