Hugging Face's logo

Spaces:
ECCV2022
/
bytetrack
Runtime error

App Files Community
2
bytetrack
File size: 3,923 Bytes 
7734d5b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
 
import json
import os


"""
cd datasets
mkdir -p mix_det/annotations
cp mot/annotations/val_half.json mix_det/annotations/val_half.json
cp mot/annotations/test.json mix_det/annotations/test.json
cd mix_det
ln -s ../mot/train mot_train
ln -s ../crowdhuman/CrowdHuman_train crowdhuman_train
ln -s ../crowdhuman/CrowdHuman_val crowdhuman_val
ln -s ../Cityscapes cp_train
ln -s ../ETHZ ethz_train
cd ..
"""

mot_json = json.load(open('datasets/mot/annotations/train_half.json','r'))

img_list = list()
for img in mot_json['images']:
    img['file_name'] = 'mot_train/' + img['file_name']
    img_list.append(img)

ann_list = list()
for ann in mot_json['annotations']:
    ann_list.append(ann)

video_list = mot_json['videos']
category_list = mot_json['categories']


print('mot17')

max_img = 10000
max_ann = 2000000
max_video = 10

crowdhuman_json = json.load(open('datasets/crowdhuman/annotations/train.json','r'))
img_id_count = 0
for img in crowdhuman_json['images']:
    img_id_count += 1
    img['file_name'] = 'crowdhuman_train/' + img['file_name']
    img['frame_id'] = img_id_count
    img['prev_image_id'] = img['id'] + max_img
    img['next_image_id'] = img['id'] + max_img
    img['id'] = img['id'] + max_img
    img['video_id'] = max_video
    img_list.append(img)
    
for ann in crowdhuman_json['annotations']:
    ann['id'] = ann['id'] + max_ann
    ann['image_id'] = ann['image_id'] + max_img
    ann_list.append(ann)

print('crowdhuman_train')

video_list.append({
    'id': max_video,
    'file_name': 'crowdhuman_train'
})


max_img = 30000
max_ann = 10000000

crowdhuman_val_json = json.load(open('datasets/crowdhuman/annotations/val.json','r'))
img_id_count = 0
for img in crowdhuman_val_json['images']:
    img_id_count += 1
    img['file_name'] = 'crowdhuman_val/' + img['file_name']
    img['frame_id'] = img_id_count
    img['prev_image_id'] = img['id'] + max_img
    img['next_image_id'] = img['id'] + max_img
    img['id'] = img['id'] + max_img
    img['video_id'] = max_video
    img_list.append(img)
    
for ann in crowdhuman_val_json['annotations']:
    ann['id'] = ann['id'] + max_ann
    ann['image_id'] = ann['image_id'] + max_img
    ann_list.append(ann)

print('crowdhuman_val')

video_list.append({
    'id': max_video,
    'file_name': 'crowdhuman_val'
})

max_img = 40000
max_ann = 20000000

ethz_json = json.load(open('datasets/ETHZ/annotations/train.json','r'))
img_id_count = 0
for img in ethz_json['images']:
    img_id_count += 1
    img['file_name'] = 'ethz_train/' + img['file_name'][5:]
    img['frame_id'] = img_id_count
    img['prev_image_id'] = img['id'] + max_img
    img['next_image_id'] = img['id'] + max_img
    img['id'] = img['id'] + max_img
    img['video_id'] = max_video
    img_list.append(img)
    
for ann in ethz_json['annotations']:
    ann['id'] = ann['id'] + max_ann
    ann['image_id'] = ann['image_id'] + max_img
    ann_list.append(ann)

print('ETHZ')

video_list.append({
    'id': max_video,
    'file_name': 'ethz'
})

max_img = 50000
max_ann = 25000000

cp_json = json.load(open('datasets/Cityscapes/annotations/train.json','r'))
img_id_count = 0
for img in cp_json['images']:
    img_id_count += 1
    img['file_name'] = 'cp_train/' + img['file_name'][11:]
    img['frame_id'] = img_id_count
    img['prev_image_id'] = img['id'] + max_img
    img['next_image_id'] = img['id'] + max_img
    img['id'] = img['id'] + max_img
    img['video_id'] = max_video
    img_list.append(img)
    
for ann in cp_json['annotations']:
    ann['id'] = ann['id'] + max_ann
    ann['image_id'] = ann['image_id'] + max_img
    ann_list.append(ann)

print('Cityscapes')

video_list.append({
    'id': max_video,
    'file_name': 'cityperson'
})

mix_json = dict()
mix_json['images'] = img_list
mix_json['annotations'] = ann_list
mix_json['videos'] = video_list
mix_json['categories'] = category_list
json.dump(mix_json, open('datasets/mix_det/annotations/train.json','w'))