-
Notifications
You must be signed in to change notification settings - Fork 1
/
infer_water.py
166 lines (138 loc) · 5.77 KB
/
infer_water.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
import numpy as np
import os
import torch
from PIL import Image, ImageCms
from torch.autograd import Variable
from torchvision import transforms
import torch.nn.functional as F
from config import ecssd_path, hkuis_path, pascals_path, sod_path, dutomron_path, \
davis_path, fbms_path, mcl_path, uvsd_path, visal_path, vos_path, segtrack_path, davsod_path, saving_path
from misc import check_mkdir, AvgMeter, cal_precision_recall_mae, cal_fmeasure
from utils.utils_mine import calculate_psnr, calculate_ssim
import time
from matplotlib import pyplot as plt
from underwater_model.model_SPOS import Water
from skimage import img_as_ubyte
import cv2
import random
torch.manual_seed(2020)
# set which gpu to use
device_id = 0
torch.cuda.set_device(device_id)
# the following two args specify the location of the file of trained model (pth extension)
# you should have the pth file in the folder './$ckpt_path$/$exp_name$'
ckpt_path = saving_path
exp_name = 'WaterEnhance_2022-06-07 08:37:34'
args = {
'snapshot': '200000', # your snapshot filename (exclude extension name)
'crf_refine': False, # whether to use crf to refine results
'save_results': True, # whether to save the resulting masks
'en_channels': [64, 128, 256],
'de_channels': 128,
'dim': 48,
# 'input_size': (380, 380),
# 'image_path': '/mnt/hdd/data/ty2/input_test',
# 'depth_path': '/mnt/hdd/data/ty2/depth_test',
# 'gt_path': '/mnt/hdd/data/ty2/gt_test',
'image_path': 'dataset image path',
'gt_path': 'dataset gt path',
'dataset': 'dataset name',
'start': 0
}
# 3, 6, 6, 5, 0, 9, 9, 1, 3, 6, 6, 1 underwater
img_transform = transforms.Compose([
transforms.ToTensor()
])
to_pil = transforms.ToPILImage()
def read_testset(dataset, image_path):
if dataset == 'UIEB':
images = os.listdir(image_path)
uieb = []
for img in images:
if img.find('deep') > 0:
continue
uieb.append(img[:-4])
return uieb
elif dataset == 'LSUI':
images = os.listdir(image_path)
lsui = []
# random_list = random.sample(range(0, len(images)), 504)
for img in images:
lsui.append(img[:-4])
return lsui
else:
images = os.listdir(image_path)
s1000 = []
for img in images:
s1000.append(img[:-4])
return s1000
def main(snapshot):
# net = R3Net(motion='', se_layer=False, dilation=False, basic_model='resnet50')
net = Water(dim=args['dim'])
# net = warp()
if snapshot is None:
print ('load snapshot \'%s\' for testing' % args['snapshot'])
net.load_state_dict(torch.load(os.path.join(ckpt_path, exp_name, args['snapshot'] + '.pth'),
map_location='cuda:' + str(device_id)))
else:
print('load snapshot \'%s\' for testing' % snapshot)
net.load_state_dict(torch.load(os.path.join(ckpt_path, exp_name, snapshot + '.pth'),
map_location='cuda:' + str(device_id)))
net.eval()
net.cuda()
results = {}
factor = 8
image_names = read_testset(args['dataset'], args['image_path'])
with torch.no_grad():
psnr_record = AvgMeter()
ssim_record = AvgMeter()
for name in image_names:
start = time.time()
img = Image.open(os.path.join(args['image_path'], name + '.jpg')).convert('RGB')
img = np.array(img)
print(img.shape)
w = img.shape[0]
h = img.shape[1]
# img = cv2.resize(img, (256, 256))
# hsv = cv2.cvtColor(img, cv2.COLOR_RGB2HSV)
lab = cv2.cvtColor(img, cv2.COLOR_RGB2LAB)
img_var = Variable(img_transform(img).unsqueeze(0), volatile=True).cuda()
lab_var = Variable(img_transform(lab).unsqueeze(0), volatile=True).cuda()
h, w = img_var.shape[2], img_var.shape[3]
H, W = ((h + factor) // factor) * factor, ((w + factor) // factor) * factor
padh = H - h if h % factor != 0 else 0
padw = W - w if w % factor != 0 else 0
img_var = F.pad(img_var, (0, padw, 0, padh), 'reflect')
lab_var = F.pad(lab_var, (0, padw, 0, padh), 'reflect')
prediction, _, _ = net(img_var, lab_var, [2, 5, 5, 4, 0, 8, 8, 1, 2, 5, 5, 1])
prediction = prediction[:, :, :h, :w]
prediction = torch.clamp(prediction, 0, 1)
prediction = prediction.permute(0, 2, 3, 1).cpu().detach().numpy()
prediction = np.squeeze(prediction)
gt = Image.open(os.path.join(args['gt_path'], name + '.jpg')).convert('RGB')
gt = np.asarray(gt)
psnr = calculate_psnr(prediction * 255.0, gt)
ssim = calculate_ssim(prediction * 255.0, gt)
if args['save_results']:
save_path = os.path.join(ckpt_path, exp_name, '%s' % (args['snapshot']), args['dataset'])
if not os.path.exists(save_path):
os.makedirs(save_path)
prediction = img_as_ubyte(prediction)
cv2.imwrite(os.path.join(save_path, name + '.png'), cv2.cvtColor(prediction, cv2.COLOR_RGB2BGR))
psnr_record.update(psnr)
ssim_record.update(ssim)
results[args['dataset']] = {'PSNR': psnr_record.avg, 'SSIM': ssim_record.avg}
print ('test results:')
print (results)
log_path = os.path.join('result_water_all.txt')
if snapshot is None:
open(log_path, 'a').write(exp_name + ' ' + args['snapshot'] + '\n')
else:
open(log_path, 'a').write(exp_name + ' ' + snapshot + '\n')
open(log_path, 'a').write(str(results) + '\n\n')
if __name__ == '__main__':
if args['start'] > 0:
for i in range(args['start'], 204000, 4000):
main(str(i))
else:
main(None)