-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathLPIPS.py
172 lines (159 loc) · 9.13 KB
/
LPIPS.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
import lpips
import glob
import torch
import os
from PIL import Image
import numpy as np
import matplotlib.pyplot as plt
class util_of_lpips():
def __init__(self, net='vgg', use_gpu=True):
'''
Parameters
----------
net: str
抽取特征的网络，['alex', 'vgg']
use_gpu: bool
是否使用GPU，默认不使用
Returns
-------
References
-------
https://github.com/richzhang/PerceptualSimilarity/blob/master/lpips_2imgs.py
'''
## Initializing the model
self.loss_fn = lpips.LPIPS(net=net)
self.use_gpu = use_gpu
if use_gpu:
self.loss_fn.cuda()
def calc_lpips(self, img1_path, img2_path):
'''
Parameters
----------
img1_path : str
图像1的路径.
img2_path : str
图像2的路径.
Returns
-------
dist01 : torch.Tensor
学习的感知图像块相似度(Learned Perceptual Image Patch Similarity, LPIPS).
References
-------
https://github.com/richzhang/PerceptualSimilarity/blob/master/lpips_2imgs.py
'''
# Load images
img0 = lpips.load_image(img1_path)
img1 = lpips.load_image(img2_path)
[h, w, c] = img1.shape
img0 = img0[:h, :w, :]
img0 = lpips.im2tensor(img0) # RGB image from [-1,1]
img1 = lpips.im2tensor(img1)
if self.use_gpu:
img0 = img0.cuda()
img1 = img1.cuda()
dist01 = self.loss_fn.forward(img0, img1)
return dist01
if __name__ == '__main__':
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/Retinex_result/test_LOL_DE/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/Retinex_result/test_LOL_HSV3/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/Retinex_result/test_LOL_HSV6/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/Retinex_result_RGB/test_LOL1/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/RetinexNet-master/test_results/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/KinD-master/results/LOLdataset_eval15/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/KinD_plus-master/test_results/LOLdataset/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/RUAS/result/LOL/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/RUAS-main/result/LOL/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/Zero-DCE-master/Zero-DCE_code/result/LOL/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/Zero-DCE_extension-main/Zero-DCE++/result/LOL/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/Retinex_result/ASWITCH/LOL/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/URetinex-Net-main/result/LOL/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/EnlightenGAN-master/result/LOL/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/home/www/myRetinex/data/LOL/eval15/low/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/Retinex_result_MIT5K/test_MIT2/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/Retinex_result_MIT5K/test_MITD2/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/Retinex_result_MIT5K/test_MIT_ND/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/Retinex_result_MIT5K/test_MIT_D/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/Retinex_result_MIT5K_RGB/test_MITND/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/RetinexNet-master/test_MIT_results/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/KinD-master/results/MIT/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/KinD_plus-master/test_results/MIT/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/RUAS/result/MIT_UPE/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/RUAS-main/result/MIT/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/Zero-DCE-master/Zero-DCE_code/result/MIT/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/Zero-DCE_extension-main/Zero-DCE++/result/MIT/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/Retinex_result/ASWITCH/MIT/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/URetinex-Net-main/result/MIT/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/EnlightenGAN-master/result/MIT/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/EXTERNAL_USB/MIT5K/test/low/']
# eval_floder = ['/media/www/EXTERNAL_USB/MIT5K/test/high/',
# '/media/www/14F492BBF4929F14/Retinex_result_RGB/test_LOL1/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/Retinex_result_MIT5K/test_LOL/']
# eval_floder = ['/home/www/myRetinex/data/LOL/eval15/high/',
# '/home/www/fsdownload/KinD-plus_results/LOL1/']
# eval_floder = ['/home/www/myRetinex/data/MIT/high/',
# '/home/www/fsdownload/RUAS-plus_result/MIT_lol/']
eval_floder = ['/media/www/14F492BBF4929F14/data/MIT/high/',
'/media/www/14F492BBF4929F14/TENR/Result/test/train_gd_num2_iter10_mit_0.85/']
# eval_floder = ['/media/www/14F492BBF4929F14/data/LOL/eval15/high/',
# '/media/www/14F492BBF4929F14/lixiaofang/study/code/Retinex/CUE-master/results/LOL/']
Lpips = util_of_lpips()
eval_data = glob.glob(eval_floder[0] + '*.*')
eval_data.sort()
num = len(eval_data)
LPIPS = []
with torch.no_grad():
for path_mat in eval_data:
name, format = os.path.basename(path_mat).split('.')
print('validating image:', name)
# gt_path = np.asarray(Image.open(path_mat))
# eIm_path = np.asarray(Image.open(eval_floder[1] + name + '.' + format))
gt_path = os.path.join(path_mat)
# eIm_path = os.path.join(eval_floder[1] + name + '.' + format)
eIm_path = os.path.join(eval_floder[1] + name + '.png')
LPIPS_enhance = Lpips.calc_lpips(gt_path, eIm_path)
print("eval image: %s, LPIPS = %5.4f" % (name, LPIPS_enhance.cpu().numpy()))
LPIPS.append(LPIPS_enhance.cpu().numpy())
# with open('/home/www/matlab_code/Retinex/metrics-MIT/our_wD.txt', 'a') as f:
with open(eval_floder[1] + 'metric.txt', 'a') as f:
# with open('/media/www/14F492BBF4929F14/Retinex_result_MIT5K/test_LOL/metric.txt', 'a') as f:
f.write("eval image: %s, LPIPS = %5.4f" % (name, LPIPS_enhance.cpu().numpy()))
f.write("\n")
avg_LPIPS = np.mean(np.asarray(LPIPS))
print("avg_LPIPS = %5.4f" % avg_LPIPS)
# with open('/home/www/matlab_code/Retinex/metrics-MIT/our_wD.txt', 'a') as f:
with open(eval_floder[1] + 'metric.txt', 'a') as f:
# with open('/media/www/14F492BBF4929F14/Retinex_result_MIT5K/test_LOL//metric.txt', 'a') as f:
f.write("avg_LPIPS = %5.4f" % avg_LPIPS)
f.write("\n")