-
Notifications
You must be signed in to change notification settings - Fork 0
/
Demo_dual.py
275 lines (199 loc) · 8.52 KB
/
Demo_dual.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
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
"""
Make Demo video
"""
import cv2
import os
import time
import sys
import cv2
import DataProvider
import numpy as np
from calcHistogram import calcHistogram3D
from Demo_2_saveBox import TrackMod_dual, TrackMod, DET_THRESHOLD, MIN_DET_THRESHOLD
demo_mode = "VIDEO"
# demo_mode = "CAM"
VIDEO_PATH = './testSample/video/lab02-18-2p.mp4'
#VIDEO_PATH = './testVideo/test_a-01-06-2p.mp4'
#VIDEO_PATH = './testVideo/test_a-01-07-2p.mp4'
#VIDEO_PATH = './testVideo/test_a-01-09-2p.mp4'#OK
#VIDEO_PATH = './testVideo/test_b-01-01-2p.mp4'#One mistake
#VIDEO_PATH = './testVideo/test_b-02-02-2p.mp4'#Worst case
#VIDEO_PATH = './testVideo/test_b-03-02-2p.mp4'#OK
#VIDEO_PATH = './testVideo/test_a-05-10-2p.mp4'#Problem
VIDEO_PATH = './testVideo/test_a-02-03-2p.mp4'#BEST
def draw_panel_dual(img_old, img_new, profile_list):
"""
Draw panel including bbox result and profile with two tracker
"""
# draw img
panel_img = np.zeros(shape=(1000,1500, 3), dtype=np.uint8)
img_old = cv2.resize(img_old, (640,480))
img_new = cv2.resize(img_new, (640,480))
h1, w1, c1 = img_new.shape
cv2.putText(panel_img, "New", (40,15), cv2.FONT_HERSHEY_COMPLEX, 0.4,(255, 255, 255), 1)
cv2.putText(panel_img, "Old", (w1+40,15), cv2.FONT_HERSHEY_COMPLEX, 0.4, (255, 255, 255), 1)
panel_img[20:20+h1, 20:20 + w1, :] = img_new
h2, w2, c2 = img_old.shape
panel_img[20:20+h2, 20+w1+20:20+w1+20+w2, :] = img_old
sub_img_size = (64, 64)
for idx, each_profile in enumerate(profile_list):
y_offset = 50 + img_new.shape[0] + idx * sub_img_size[1]
x_init_offset = 20
msg_id = str(each_profile.uid)
# current saved image
cv2.putText(panel_img, "Profile " + msg_id, (x_init_offset, y_offset+32), cv2.FONT_HERSHEY_COMPLEX, 0.4, (255, 255, 255), 1)
curr_frame_max = 10
for idx in range(curr_frame_max):
len_profile = len(each_profile.rect_image_list)
if idx >= len_profile:
break
if len_profile < curr_frame_max:
curr_idx = idx
else:
curr_idx = len_profile - curr_frame_max + idx
each_positive_img_rect = each_profile.rect_image_list[curr_idx]
curr_img, curr_rect = each_positive_img_rect
curr_cropped_img = curr_img[curr_rect[1]:curr_rect[1] + curr_rect[3],
curr_rect[0]:curr_rect[0] + curr_rect[2]]
curr_cropped_img = cv2.resize(curr_cropped_img, sub_img_size)
x_offset = x_init_offset + sub_img_size[0] * (idx + 1)
panel_img[y_offset:y_offset + sub_img_size[1], x_offset:x_offset + sub_img_size[0], :] = curr_cropped_img
return panel_img
def draw_panel(img, profile_list):
"""
Draw panel including bbox result and profile
"""
# draw img
panel_img = np.zeros(shape=(900, 1500, 3), dtype=np.uint8)
h1, w1, c1 = img.shape
panel_img[20:20 + h1, 20:20 + w1, :] = img
sub_img_size = (64, 64)
for idx, each_profile in enumerate(profile_list):
y_offset = 20 + img.shape[0] + idx * sub_img_size[1]
x_init_offset = 20
msg_id = str(each_profile.uid)
# current saved image
cv2.putText(panel_img, msg_id, (x_init_offset, y_offset), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 255, 255), 1)
curr_frame_max = 10
for idx in range(curr_frame_max):
len_profile = len(each_profile.rect_image_list)
if idx >= len_profile:
break
if len_profile < curr_frame_max:
curr_idx = idx
else:
curr_idx = len_profile - curr_frame_max + idx
each_positive_img_rect = each_profile.rect_image_list[curr_idx]
curr_img, curr_rect = each_positive_img_rect
curr_cropped_img = curr_img[curr_rect[1]:curr_rect[1] + curr_rect[3],
curr_rect[0]:curr_rect[0] + curr_rect[2]]
curr_cropped_img = cv2.resize(curr_cropped_img, sub_img_size)
x_offset = x_init_offset + sub_img_size[0] * (idx + 1)
panel_img[y_offset:y_offset + sub_img_size[1], x_offset:x_offset + sub_img_size[0], :] = curr_cropped_img
return panel_img
def demo_total(video_path):
if demo_mode == "VIDEO":
dp = DataProvider.VideoDataProvider(video_path)
resize_rate = 0.4
tracker = TrackMod_dual(conf_file="cfg/yolo-f.cfg", model_file=8000, det_threshold=DET_THRESHOLD,
min_det_treshold=MIN_DET_THRESHOLD,
use_cmatch=True, tracker_type='TM', tracker_limit=10)
frame_idx = 0
while True:
img = dp.get()
if img is None:
break
img_resize = cv2.resize(img, (0, 0), fx=resize_rate, fy=resize_rate)
img_resize_draw_old = img_resize.copy()
img_resize_draw_new = img_resize.copy()
tracker.run(img_resize)
tracker.draw_old(img_resize_draw_old)
tracker.draw_new(img_resize_draw_new)
"""
if frame_idx > 572:
panel = draw_panel(img_resize_draw, tracker.tracking.profile_classifier.profile_list)
cv2.imshow("Tracking Result", panel)
cv2.waitKey()
"""
panel = draw_panel_dual(img_resize_draw_old, img_resize_draw_new, tracker.tracking_new.profile_classifier.profile_list)
cv2.imshow("Tracking Result", panel)
print(frame_idx)
print('New')
for trackRes in tracker.current_tracks_new:
id = trackRes.id
x1 = trackRes.tl[0]
y1 = trackRes.tl[1]
x2 = trackRes.br[0]
y2 = trackRes.br[1]
last_state = trackRes.last_state
type = trackRes.type
print(str(type) + ',' + str(id) + ',' + last_state + ',' + str(x1) + ',' + str(x2) + ',' + str(y1) + ',' + str(y2))
print('Old')
for trackRes in tracker.current_tracks_old:
id = trackRes.id
x1 = trackRes.tl[0]
y1 = trackRes.tl[1]
x2 = trackRes.br[0]
y2 = trackRes.br[1]
last_state = trackRes.last_state
type = trackRes.type
print(str(type) + ',' + str(id) + ',' + last_state + ',' + str(x1) + ',' + str(x2) + ',' + str(
y1) + ',' + str(y2))
frame_idx += 1
key = cv2.waitKey(1)
if key == ord('q'):
break
if key == ord('c'):
cv2.waitKey()
def demo_single(video_path, old_tracker=False):
if demo_mode == "VIDEO":
dp = DataProvider.VideoDataProvider(video_path)
resize_rate = 0.4
if old_tracker:
tracker = TrackMod(conf_file="cfg/yolo-f.cfg", model_file=8000, det_threshold=DET_THRESHOLD,
min_det_treshold=MIN_DET_THRESHOLD,
use_cmatch=True, tracker_type='TM', tracker_limit=10, old_tracker=True)
else:
tracker = TrackMod(conf_file="cfg/yolo-f.cfg", model_file=8000, det_threshold=DET_THRESHOLD,
min_det_treshold=MIN_DET_THRESHOLD,
use_cmatch=True, tracker_type='TM', tracker_limit=10, old_tracker=False)
frame_idx = 0
while True:
img = dp.get()
if img is None:
break
img_resize = cv2.resize(img, (0, 0), fx=resize_rate, fy=resize_rate)
img_resize_draw = img_resize.copy()
tracker.run(img_resize)
tracker.draw(img_resize_draw)
"""
if frame_idx > 572:
panel = draw_panel(img_resize_draw, tracker.tracking.profile_classifier.profile_list)
cv2.imshow("Tracking Result", panel)
cv2.waitKey()
"""
if old_tracker:
panel = draw_panel(img_resize_draw, [])
else:
panel = draw_panel(img_resize_draw, tracker.tracking.profile_classifier.profile_list)
cv2.imshow("Tracking Result", panel)
print(frame_idx)
for trackRes in tracker.current_tracks:
id = trackRes.id
x1 = trackRes.tl[0]
y1 = trackRes.tl[1]
x2 = trackRes.br[0]
y2 = trackRes.br[1]
last_state = trackRes.last_state
type = trackRes.type
print(str(type) + ',' + str(id) + ',' + last_state + ',' + str(x1) + ',' + str(x2) + ',' + str(y1) + ',' + str(y2))
frame_idx += 1
key = cv2.waitKey(1)
print(key)
if key == ord('q'):
break
if key == ord('c'):
cv2.waitKey()
if __name__ == "__main__":
demo_total(VIDEO_PATH)
#demo_single(VIDEO_PATH, old_tracker=True)