-
Notifications
You must be signed in to change notification settings - Fork 0
/
openposeto_cocojson_headonly.py
226 lines (189 loc) · 7.09 KB
/
openposeto_cocojson_headonly.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
# import numpy as np
# import pandas as pd
# import cv2
# from pycococreatortools import *
import os, sys, shutil
# import time
import datetime
import json
import argparse
import cv2
def get_image_info(image_name):
file_name = image_name
id = image_name.split('.')[0]
#id = int(id)
return id, file_name
def get_segmentation_info(seg_path):
with open(seg_path) as f:
json_data = json.load(f)
shapes = json_data['shapes']
segmentations = []
for i in range(len(shapes)):
segmentation = []
point = shapes[i]['points']
for j in range(len(point)):
segmentation.append(point[j][0])
segmentation.append(point[j][1])
segmentations.append(segmentation)
return segmentations
def get_keypoints_info(json_path):
with open(json_path) as f:
json_data = json.load(f)
pdata = json_data['people']
pose_keypoints = pdata['pose_keypoints_2d']
num_keypoints = 0
keypoints = pose_keypoints
# for i in range(3):
# keypoints.pop(3)
for i in range(0, len(keypoints), 3):
if keypoints[i] != 0 or keypoints[i + 1] != 0:
keypoints[i + 2] = 2
num_keypoints += 1
else:
keypoints[i] = 0
keypoints[i + 1] = 0
keypoints[i + 2] = 0
print(num_keypoints)
print(len(keypoints))
return keypoints, num_keypoints
def create_image_info(image_id, file_name, image_size,
date_captured=datetime.datetime.utcnow().isoformat(' '),
license_id=1, coco_url="", flickr_url=""):
image = {
"id": image_id,
"file_name": file_name,
"width": image_size[0],
"height": image_size[1],
"date_captured": date_captured,
"license": license_id,
"coco_url": coco_url,
"flickr_url": flickr_url
}
return image
def create_annotation_info(num_keypoints, keypoints, image_id, id):
annotation = {
"segmentation": [
[125.12, 539.69, 140.94, 522.43, 100.67, 496.54, 84.85, 469.21,
73.35, 450.52, 104.99, 342.65, 168.27, 290.88, 179.78, 288]],
"num_keypoints": num_keypoints,
"area": 47803.27955,
"iscrowd": 0,
"keypoints": keypoints,
"image_id": image_id,
"bbox": [73.35, 206.02, 300.58, 372.5],
"category_id": 1,
"id": id
}
return annotation
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument("--images_dir_name", type=str, default='/home/dabai/datasets/aichallenger/tmp',required=False,
help="image's file")
parser.add_argument("--jsons_dir_name", type=str, default='/home/dabai/datasets/aichallenger/ai_challenger_json/',required=False,
help="jsons path")
parser.add_argument("--out_json_path", type=str, default = './rrr.json',required=False,
help="openpose 2 coco path")
args = parser.parse_args()
coco_output = {}
# size = [1920,1080]
info = {
"description": "",
"url": "",
"version": "",
"year": 2019,
"contributor": "",
"date_created": datetime.datetime.utcnow().isoformat(' ')
}
licenses = [
{
"id": 1,
"name": "",
"url": ""
}
]
categories = [
{
"supercategory": "person",
"id": 1,
"name": "person",
"keypoints": [
"nose1","right_shoulder2", "right_elbow3", "right_wrist4",
"left_shoulder5", "left_elbow6", "left_wrist7", "right_hip9",
"right_knee10", "right_ankle11", "left_hip12", "left_knee13", "left_ankle14",
"right_eye15", "left_eye16", "right_ear17", "left_ear18",
"hip8",
"right_heel24", "right_bigtoe22", "right_smalltoe23",
"left_heel21", "left_bigtoe19", "left_smalltoe20"],
"skeleton": [
# Nose, Neck, R hand, L hand, R leg, L leg, Eyes, Ears
[1, 15], [15, 17], [1, 16], [16, 18],
[2, 3], [3, 4], [2, 5], [5, 6], [6, 7],
[8, 9], [9, 10], [10, 11], [11, 23], [11, 24], [24, 22],
[8, 12], [12, 13], [13, 14], [14, 21], [14, 20], [19, 21],
[2, 17], [5, 18]
# [1, 25], [25, 8],
# [2, 9], [5, 12]
]
}
]
images = []
annotations = []
# images
images_dir = os.listdir(args.images_dir_name)
images_id = []
for image_name in images_dir:
#print(os.path.join(args.images_dir_name, image_name))
img = cv2.imread(os.path.join(args.images_dir_name, image_name))
size = [img.shape[1], img.shape[0]]
id, file_name = get_image_info(image_name=image_name)
image = create_image_info(id, file_name, size)
images.append(image)
images_id.append(id)
print("-----finished images-----")
# annotations
jsons_dir = os.listdir(args.jsons_dir_name)
id = 0
for j in range(len(jsons_dir)):
json_file = args.jsons_dir_name + jsons_dir[j]
image_id = jsons_dir[j].split('_')[0]
#image_id = int(image_id)
with open(json_file) as f:
json_data = json.load(f)
pdatas = json_data['people']
for pdata in pdatas:
num_keypoints = 0
keypoints = pdata['pose_keypoints_2d']
new_keypoints = keypoints
for j in range(3):
new_keypoints.pop(3)
for i in range(0, len(new_keypoints), 3):
# if new_keypoints[i+2] > 0.1:
if new_keypoints[i] != 0 or new_keypoints[i + 1] != 0:
new_keypoints[i + 2] = 2
num_keypoints += 1
else:
new_keypoints[i] = new_keypoints[i + 1] = new_keypoints[i + 2] = 0
id += 1
print(len(new_keypoints))
new_keypoints0 = new_keypoints[0:3]#鼻子
new_keypoints1 = new_keypoints[42:54]#眼睛,耳朵
new_keypoints1.append(new_keypoints0[0])
new_keypoints1.append(new_keypoints0[1])
new_keypoints1.append(new_keypoints0[2])
new_keypoints = new_keypoints1
print(len(new_keypoints))
annotation = create_annotation_info(num_keypoints, new_keypoints, image_id, id)
annotations.append(annotation)
#print(annotations)
print("-----finished annotations-----")
coco_output["info"] = info
coco_output["licenses"] = licenses
coco_output["categories"] = categories
coco_output["images"] = images
coco_output["annotations"] = annotations
print("transform to coco")
# coco = pd.DataFrame(coco_output)
# coco_output.to_json('./openpose.json')
with open(args.out_json_path, 'w') as f:
json.dump(coco_output, f)
print('write in json')