-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathdemo.py
168 lines (144 loc) · 5.45 KB
/
demo.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
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import argparse
import os
from datetime import datetime
import socket
import timeit
from tensorboardX import SummaryWriter
import numpy as np
import pycocotools.mask as rletools
import torch
import torch.optim as optim
from torchvision import transforms
from torch.utils.data import DataLoader
import matplotlib.pyplot as plt
import torch.nn as nn
import cv2
import torch.nn.functional as F
from network.backbone import ResNet
from network.seghead import SegHead
import dataloaders.MOTS_dataloaders as ms
import dataloaders.Demo_dataloader as de
from network.fpn import FPN101
from network.GeneralizedRCNN import GeneralizedRCNN
from torch.utils.tensorboard import SummaryWriter
def get_img_size(sequence):
if sequence==5 or sequence==6:
return [640,480]
else:
return [1920,1080]
def normalization(data):
_range = np.max(data) - np.min(data)
return (data - np.min(data)) / _range
def main(sequence):
gpu_id = 0
device = torch.device("cuda:" + str(gpu_id) if torch.cuda.is_available() else "cpu")
save_dir = "models"
if not os.path.exists(save_dir):
os.makedirs(os.path.join(save_dir))
backbone = GeneralizedRCNN()
# seghead=SegHead(get_img_size(sequence))
seghead = SegHead([2048,1024])
BackBoneName = "GeneralizedRCNN"
SegHeadName = "seghead"
backbone.load_state_dict(
torch.load(os.path.join(save_dir, BackBoneName + '_epoch-' + str(112) + '.pth'),
map_location=lambda storage, loc: storage))
seghead.load_state_dict(
torch.load(os.path.join(save_dir, SegHeadName + '_epoch-' + str(112) + '.pth'),
map_location=lambda storage, loc: storage))
backbone=backbone.cuda()
seghead=seghead.cuda()
test = de.MOTTrackDataset(sequence=sequence)
testloader = DataLoader(test)
backbone.eval()
seghead.eval()
file = open('{:04}.txt'.format(sequence), "w")
size = get_img_size(sequence)
with torch.no_grad():
for ii, sample_batched in enumerate(testloader):
inputs, bbox,track_list = sample_batched["img"],sample_batched["bbox"],sample_batched["track"]
inputs = inputs.cuda()
writer=SummaryWriter()
writer.add_graph(backbone,(inputs,))
writer.close()
feature = backbone(inputs)
output_list = []
for l in range(1):
out = seghead(feature,level=3-l)
output = out.detach().cpu().numpy()[0][0]
# plt.imshow(output)
# plt.show()
if sequence == 5 or sequence == 6:
output = cv2.resize(output, (640, 480))
else:
output = cv2.resize(output, (1920, 1080))
output_list.append(output)
output = sum(output_list)
# output = normalization(output)
output[output < 0.5] = 0
# plt.imshow(output)
# plt.show()
result_list=[]
for nbox,track_id in zip(bbox,track_list):
box = nbox.squeeze()
box = [box[0]/2048*size[0],box[1]/1024*size[1],box[2]/2048*size[0],box[3]/1024*size[1]]
box = [int(i) for i in box]
result_list.append((box,track_id))
result_list.sort(key=lambda item:item[0][3]*item[0][2])
result_list.reverse()
for item in result_list:
box = item[0]
track_id = item[1]
mask = np.zeros_like(output)
temp = mask[box[1]:box[1]+box[3],box[0]:box[0]+box[2]]
# pre=normalization(pre)
temp[temp>-1] = 1
real = output.copy()
real[mask<1]=0
output[mask>0]=0
real[real>0]=1
# plt.imshow(real)
# plt.show()
mask = np.asfortranarray(real)
mask = mask.astype(np.uint8)
rle = rletools.encode(mask)
# print(rletools.area(rle))
if rletools.area(rle) < 10:
continue
line = ' '.join([str(ii + 1), str(int(2000 + track_id)), "2", str(rle['size'][0]), str(rle['size'][1]),
rle['counts'].decode(encoding='UTF-8')])
file.write(line + '\n')
#
#
# if sequence==5 or sequence==6:
# background = cv2.resize(background, (640, 480))
# else:
# background = cv2.resize(background, (1920, 1080))
#
#
#
# for id in track_list:
# id = int(id)
# mask = np.zeros_like(background)
# mask[background==id]=1
# mask = np.asfortranarray(mask)
# mask = mask.astype(np.uint8)
# rle = rletools.encode(mask)
# # print(id,rletools.area(rle))
# if rletools.area(rle)<2000:
# continue
# output =' '.join([str(ii+1),str(int(2000+id)),"2",str(rle['size'][0]),str(rle['size'][1]),rle['counts'].decode(encoding='UTF-8')])
# file.write(output+'\n')
file.close()
if __name__ == "__main__":
main(2)
print("finish:2")
main(5)
print("finish:5")
main(9)
print("finish:9")
main(11)
print("finish:11")