-
Notifications
You must be signed in to change notification settings - Fork 111
/
Copy pathtest_frontal.py
210 lines (170 loc) · 7.18 KB
/
test_frontal.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
import torch.multiprocessing as multiprocessing
multiprocessing.set_start_method('spawn', force=True)
from models.networks.sync_batchnorm import DataParallelWithCallback
import sys
import numpy as np
import os
import data
from util.iter_counter import IterationCounter
from options.test_options import TestOptions
from models.test_model import TestModel
from util.visualizer import Visualizer
from util import html, util
from torch.multiprocessing import Process, Queue, Pool
from data.data_utils import init_parallel_jobs
from skimage import transform as trans
import cv2
import time
import torch
from models.networks.test_render import TestRender
def create_path(a_path, b_path):
name_id_path = os.path.join(a_path, b_path)
if not os.path.exists(name_id_path):
os.makedirs(name_id_path)
return name_id_path
def create_paths(save_path, img_path, foldername='orig', folderlevel=2):
save_rotated_path_name = create_path(save_path, foldername)
path_split = img_path.split('/')
rotated_file_savepath = save_rotated_path_name
for level in range(len(path_split) - folderlevel, len(path_split)):
file_name = path_split[level]
rotated_file_savepath = os.path.join(rotated_file_savepath, file_name)
return rotated_file_savepath
def affine_align(img, landmark=None, **kwargs):
M = None
src = np.array([
[38.2946, 51.6963],
[73.5318, 51.5014],
[56.0252, 71.7366],
[41.5493, 92.3655],
[70.7299, 92.2041] ], dtype=np.float32 )
src=src * 224 / 112
dst = landmark.astype(np.float32)
tform = trans.SimilarityTransform()
tform.estimate(dst, src)
M = tform.params[0:2,:]
warped = cv2.warpAffine(img, M, (224, 224), borderValue = 0.0)
return warped
def landmark_68_to_5(t68):
le = t68[36:42, :].mean(axis=0, keepdims=True)
re = t68[42:48, :].mean(axis=0, keepdims=True)
no = t68[31:32, :]
lm = t68[48:49, :]
rm = t68[54:55, :]
t5 = np.concatenate([le, re, no, lm, rm], axis=0)
t5 = t5.reshape(10)
return t5
def save_img(img, save_path):
image_numpy = util.tensor2im(img)
util.save_image(image_numpy, save_path, create_dir=True)
return image_numpy
if __name__ == '__main__':
opt = TestOptions().parse()
data_info = data.dataset_info()
datanum = data_info.get_dataset(opt)
folderlevel = data_info.folder_level[datanum]
dataloaders = data.create_dataloader_test(opt)
visualizer = Visualizer(opt)
iter_counter = IterationCounter(opt, len(dataloaders[0]) * opt.render_thread)
# create a webpage that summarizes the all results
testing_queue = Queue(10)
ngpus = opt.device_count
render_gpu_ids = list(range(ngpus - opt.render_thread, ngpus))
render_layer_list = []
for gpu in render_gpu_ids:
opt.gpu_ids = gpu
render_layer = TestRender(opt)
render_layer_list.append(render_layer)
opt.gpu_ids = list(range(0, ngpus - opt.render_thread))
print('Testing gpu ', opt.gpu_ids)
if opt.names is None:
model = TestModel(opt)
model.eval()
model = torch.nn.DataParallel(model.cuda(),
device_ids=opt.gpu_ids,
output_device=opt.gpu_ids[-1],
)
models = [model]
names = [opt.name]
save_path = create_path(create_path(opt.save_path, opt.name), opt.dataset)
save_paths = [save_path]
f = [open(
os.path.join(save_path, opt.dataset + str(opt.list_start) + str(opt.list_end) + '_rotate_lmk.txt'), 'w')]
else:
models = []
names = []
save_paths = []
f = []
for name in opt.names.split(','):
opt.name = name
model = TestModel(opt)
model.eval()
model = torch.nn.DataParallel(model.cuda(),
device_ids=opt.gpu_ids,
output_device=opt.gpu_ids[-1],
)
models.append(model)
names.append(name)
save_path = create_path(create_path(opt.save_path, opt.name), opt.dataset)
save_paths.append(save_path)
f_rotated = open(
os.path.join(save_path, opt.dataset + str(opt.list_start) + str(opt.list_end) + '_rotate_lmk.txt'), 'w')
f.append(f_rotated)
test_tasks = init_parallel_jobs(testing_queue, dataloaders, iter_counter, opt, render_layer_list)
# test
landmarks = []
process_num = opt.list_start
first_time = time.time()
try:
for i, data_i in enumerate(range(len(dataloaders[0]) * opt.render_thread)):
# if i * opt.batchSize >= opt.how_many:
# break
# data = trainer.get_input(data_i)
start_time = time.time()
data = testing_queue.get(block=True)
current_time = time.time()
time_per_iter = (current_time - start_time) / opt.batchSize
message = '(************* each image render time: %.3f *****************) ' % (time_per_iter)
print(message)
img_path = data['path']
rotated_landmarks = data['rotated_landmarks'][:, :, :2].cpu().numpy().astype(np.float)
generate_rotateds = []
for model in models:
generate_rotated = model.forward(data, mode='single')
generate_rotateds.append(generate_rotated)
for n, name in enumerate(names):
opt.name = name
for b in range(generate_rotateds[n].shape[0]):
# get 5 key points
rotated_keypoints = landmark_68_to_5(rotated_landmarks[b])
# get savepaths
rotated_file_savepath = create_paths(save_paths[n], img_path[b], folderlevel=folderlevel)
image_numpy = save_img(generate_rotateds[n][b], rotated_file_savepath)
rotated_keypoints_str = rotated_file_savepath + ' 1 ' + ' '.join([str(int(n)) for n in rotated_keypoints]) + '\n'
print('process image...' + rotated_file_savepath)
f[n].write(rotated_keypoints_str)
current_time = time.time()
if n == 0:
process_num += 1
print('processed num ' + str(process_num))
if opt.align:
aligned_file_savepath = create_paths(save_paths[n], img_path[b], 'aligned', folderlevel=folderlevel)
warped = affine_align(image_numpy, rotated_keypoints.reshape(5, 2))
util.save_image(warped, aligned_file_savepath, create_dir=True)
current_time = time.time()
time_per_iter = (current_time - start_time) / opt.batchSize
message = '(************* each image time total: %.3f *****************) ' % (time_per_iter)
print(message)
except KeyboardInterrupt:
print("Interrupted!")
for fs in f:
fs.close()
pass
except Exception as e:
print(e)
for fs in f:
fs.close()
else:
print('finished')
for fs in f:
fs.close()