/
resume_from_coco.py
228 lines (197 loc) · 8.06 KB
/
resume_from_coco.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
215
216
217
218
219
220
221
222
223
224
225
226
227
228
from __future__ import print_function
import os
import argparse
import torch
import torch.backends.cudnn as cudnn
import numpy as np
from torch.autograd import Variable
from data import BaseTransform, VOC_300, VOC_512, COCO_300, COCO_512, COCO_mobile_300
from data import VOC_CLASSES as labelmap
from layers.functions import Detect, PriorBox
from utils.timer import Timer
import cv2
parser = argparse.ArgumentParser(description='Receptive Field Block Net')
parser.add_argument('-v', '--version', default='SSD_mobile',
help='RFB_vgg ,RFB_E_vgg or RFB_mobile version.')
parser.add_argument('-s', '--size', default='300',
help='300 or 512 input size.')
parser.add_argument('-d', '--dataset', default='VOC',
help='VOC or COCO version')
parser.add_argument('-m', '--trained_model',
default='/Users/fotoable/workplace/pytorch_ssd/weights/SSD_mobile_COCO_epoches_150.pth',
type=str, help='Trained state_dict file path to open')
parser.add_argument('--save_folder', default='eval/', type=str,
help='Dir to save results')
parser.add_argument('--cuda', default=False, type=bool,
help='Use cuda to train model')
parser.add_argument('--retest', default=False, type=bool,
help='test cache results')
args = parser.parse_args()
if not os.path.exists(args.save_folder):
os.mkdir(args.save_folder)
if args.dataset == 'VOC':
cfg = (VOC_300, VOC_512)[args.size == '512']
else:
cfg = (COCO_300, COCO_512)[args.size == '512']
if args.version == 'RFB_vgg':
from models.RFB_Net_vgg import build_net
elif args.version == 'RFB_E_vgg':
from models.RFB_Net_E_vgg import build_net
elif args.version == 'RFB_mobile':
from models.RFB_Net_mobile import build_net
cfg = COCO_mobile_300
elif args.version == 'SSD_vgg':
from models.SSD_vgg import build_net
elif args.version == 'FSSD_vgg':
from models.FSSD_vgg import build_net
elif args.version == 'SSD_mobile':
from models.SSD_mobile import build_net
cfg = COCO_mobile_300
else:
print('Unkown version!')
priorbox = PriorBox(cfg)
priors = Variable(priorbox.forward(), volatile=True)
def py_cpu_nms(dets, thresh):
"""Pure Python NMS baseline."""
x1 = dets[:, 0]
y1 = dets[:, 1]
x2 = dets[:, 2]
y2 = dets[:, 3]
scores = dets[:, 4]
areas = (x2 - x1 + 1) * (y2 - y1 + 1)
order = scores.argsort()[::-1]
keep = []
while order.size > 0:
i = order[0]
keep.append(i)
xx1 = np.maximum(x1[i], x1[order[1:]])
yy1 = np.maximum(y1[i], y1[order[1:]])
xx2 = np.minimum(x2[i], x2[order[1:]])
yy2 = np.minimum(y2[i], y2[order[1:]])
w = np.maximum(0.0, xx2 - xx1 + 1)
h = np.maximum(0.0, yy2 - yy1 + 1)
inter = w * h
ovr = inter / (areas[i] + areas[order[1:]] - inter)
inds = np.where(ovr <= thresh)[0]
order = order[inds + 1]
return keep
class ObjectDetector:
def __init__(self, net, detection, transform, num_classes=21, cuda=False, max_per_image=300, thresh=0.5):
self.net = net
self.detection = detection
self.transform = transform
self.max_per_image = 300
self.num_classes = num_classes
self.max_per_image = max_per_image
self.cuda = cuda
self.thresh = thresh
def predict(self, img):
scale = torch.Tensor([img.shape[1], img.shape[0],
img.shape[1], img.shape[0]]).cpu().numpy()
_t = {'im_detect': Timer(), 'misc': Timer()}
assert img.shape[2] == 3
x = Variable(self.transform(img).unsqueeze(0), volatile=True)
if self.cuda:
x = x.cuda()
_t['im_detect'].tic()
out = net(x, test=True) # forward pass
boxes, scores = self.detection.forward(out, priors)
detect_time = _t['im_detect'].toc()
boxes = boxes[0]
scores = scores[0]
boxes = boxes.cpu().numpy()
scores = scores.cpu().numpy()
# scale each detection back up to the image
boxes *= scale
_t['misc'].tic()
all_boxes = [[] for _ in range(num_classes)]
for j in range(1, num_classes):
inds = np.where(scores[:, j] > self.thresh)[0]
if len(inds) == 0:
all_boxes[j] = np.zeros([0, 5], dtype=np.float32)
continue
c_bboxes = boxes[inds]
c_scores = scores[inds, j]
print(scores[:, j])
c_dets = np.hstack((c_bboxes, c_scores[:, np.newaxis])).astype(
np.float32, copy=False)
# keep = nms(c_bboxes,c_scores)
keep = py_cpu_nms(c_dets, 0.45)
keep = keep[:50]
c_dets = c_dets[keep, :]
all_boxes[j] = c_dets
if self.max_per_image > 0:
image_scores = np.hstack([all_boxes[j][:, -1] for j in range(1, num_classes)])
if len(image_scores) > self.max_per_image:
image_thresh = np.sort(image_scores)[-self.max_per_image]
for j in range(1, num_classes):
keep = np.where(all_boxes[j][:, -1] >= image_thresh)[0]
all_boxes[j] = all_boxes[j][keep, :]
nms_time = _t['misc'].toc()
print('net time: ', detect_time)
print('post time: ', nms_time)
return all_boxes
COLORS = [(255, 0, 0), (0, 255, 0), (0, 0, 255)]
FONT = cv2.FONT_HERSHEY_SIMPLEX
def get_coco_voc_mask(class_num, select_index, prior_num):
tmp_mask = []
prior_num = int(prior_num)
for i in range(prior_num):
for j in range(len(select_index)):
tmp_mask.append(class_num * i + select_index[j])
coco_voc_mask = torch.LongTensor(tmp_mask)
return coco_voc_mask
if __name__ == '__main__':
# load net
coco_voc_list = []
with open('coco_voc.txt') as label_file:
coco_voc_list = label_file.readlines()
coco_voc_list = [int(item.split(',')[0]) for item in coco_voc_list]
img_dim = (300, 512)[args.size == '512']
num_classes = (21, 81)[args.dataset == 'COCO']
net = build_net(img_dim, num_classes) # initialize detector
state_dict = torch.load(args.trained_model, map_location=lambda storage, loc: storage)
select_index = coco_voc_list
# print(state_dict['module.conf.0.bias'])
# create new OrderedDict that does not contain `module.`
from collections import OrderedDict
new_state_dict = OrderedDict()
for k, v in state_dict.items():
head = k[:7]
if 'conf' in k:
prior_num = v.size(0) / 81
coco_voc_mask = get_coco_voc_mask(81, select_index, prior_num)
v = torch.index_select(v, 0, coco_voc_mask)
if head == 'module.':
name = k[7:] # remove `module.`
else:
name = k
new_state_dict[name] = v
net.load_state_dict(new_state_dict)
torch.save(net.state_dict(), 'ssd_mobile_net_coco_voc.pth')
net.eval()
print('Finished loading model!')
print(net)
# load data
if args.cuda:
net = net.cuda()
cudnn.benchmark = True
# evaluation
top_k = (300, 200)[args.dataset == 'COCO']
detector = Detect(num_classes, 0, cfg)
rgb_means = ((104, 117, 123), (103.94, 116.78, 123.68))[args.version == 'RFB_mobile']
transform = BaseTransform(net.size, rgb_means, (2, 0, 1))
object_detector = ObjectDetector(net, detector, transform, num_classes)
image = cv2.imread('test.jpg')
detect_bboxes = object_detector.predict(image)
for class_id, class_collection in enumerate(detect_bboxes):
if len(class_collection) > 0:
for i in range(class_collection.shape[0]):
if class_collection[i, -1] > 0.6:
pt = class_collection[i]
cv2.rectangle(image, (int(pt[0]), int(pt[1])), (int(pt[2]),
int(pt[3])), COLORS[i % 3], 2)
cv2.putText(image, labelmap[class_id - 1], (int(pt[0]), int(pt[1])), FONT,
0.5, (255, 255, 255), 2)
cv2.imshow('result', image)
cv2.waitKey()