YOLOPOSE算法COCOkeypoints关键点标注YOLO格式转COCO格式，供coco annotator标注软件导入_yolov8-pose的关键点数据集转为coco格式

import json
import os

import cv2
import numpy as np
import random

def randomcolor():
    colorArr = ['1','2','3','4','5','6','7','8','9','A','B','C','D','E','F']
    color = ""
    for i in range(6):
        color += colorArr[random.randint(0,14)]
    return "#"+color

if __name__ == '__main__':

    # 获取存放txt的文件夹
    # 迭代处理每个文件
    # 迭代处理每个文件中的每一行 每个文件插入到一个json

    txt_path = "../txt"
    # img_path = "E:\PythonWorkSpace\paperTrain\WiderPersonAll\Images"
    img_path = "../images"
    json_path = "../json"


    write_json_context = dict()  # 写入.json文件的大字典
    # write_json_context['info'] = {'description': '', 'url': '', 'version': '', 'year': 2023, 'contributor': '',
    #                               'date_created': '2023-02-16'}
    # write_json_context['licenses'] = [{'id': 1, 'name': None, 'url': None}]
    write_json_context['categories'] = [
        {"id": 1, "name": "person", "supercategory": "", "color": "#75e955", "metadata": {}, "keypoint_colors": [],
         "keypoints": ["nose", "left_eye", "right_eye", "left_ear", "right_ear", "left_shoulder", "right_shoulder",
                       "left_elbow", "right_elbow", "left_wrist", "right_wrist", "left_hip", "right_hip", "left_knee",
                       "right_knee", "left_ankle", "right_ankle"],
         "skeleton": [[16, 14], [14, 12], [17, 15], [15, 13], [12, 13], [6, 12], [7, 13], [6, 7], [6, 8], [7, 9],
                      [8, 10], [9, 11], [2, 3], [1, 2], [1, 3], [2, 4], [3, 5], [4, 6], [5, 7]]}]

    img_list = os.listdir(img_path)
    for index, img_name in enumerate(img_list):
        write_json_context['images'] = []
        write_json_context['annotations'] = []
        print(img_name)
        # f = open(txt_path+"/"+img_name)
        # 遍历一个txt的每一行
        # for line in f.readlines():
        #     det_list = list(map(float, line.split(' ')))
        #     print(det_list)
        # lines1 = f.readlines()
        # print(lines1)
        categories = []

        im0 = cv2.imread(img_path+"/"+img_name)
        W, H = im0.shape[1], im0.shape[0]

        img_context = {}  # 使用一个字典存储该图片信息
        # img_name=os.path.basename(imagePath)   #返回path最后的文件名。如果path以/或\结尾，那么就会返回空值
        img_context["file_name"] = img_name
        src_front, src_back = os.path.splitext(img_name)  # 将文件名和文件格式分开
        # print("imageFile", src_front)
        img_context["height"] = H
        img_context["width"] = W
        # img_context["date_captured"] = '2021-07-25'
        img_context["id"] = index  # 该图片的id
        # img_context["license"] = 1
        # img_context["color_url"] = ""
        # img_context["flickr_url"] = ""
        img_context["path"] = "/datasets/wider/"+img_name
        img_context["dataset_id"] = "3"
        write_json_context["images"].append(img_context)  # 将该图片信息添加到'image'列表中

        txt_name = src_front + '.txt' #对应的txt文件名
        with open(os.path.join(txt_path, txt_name), 'r') as fr:
            lines = fr.readlines()

        # 迭代一个txt文件中的每一行，插入到annotations
        for j, line in enumerate(lines):
            bbox_dict = {}

            class_id, x, y, w, h = line.strip().split(' ')[:5]  # 获取每一个标注框的详细信息
            class_id, x, y, w, h = int(class_id), float(x), float(y), float(w), float(h)  # 将字符串类型转为可计算的int和float类型

            xmin = (x - w / 2) * W  # 坐标转换
            ymin = (y - h / 2) * H
            xmax = (x + w / 2) * W
            ymax = (y + h / 2) * H
            w = w * W
            h = h * H

            bbox_dict["id"] = index * 10000 + j  # bounding box的坐标信息
            bbox_dict["image_id"] = index
            bbox_dict["category_id"] = class_id + 1  # 注意目标类别要加一
            bbox_dict["iscrowd"] = False
            bbox_dict["isbbox"] = True
            bbox_dict["color"] = randomcolor()
            height, width = abs(ymax - ymin), abs(xmax - xmin)
            bbox_dict["area"] = np.round(height * width)
            bbox_dict["bbox"] = list(np.round([xmin, ymin, w, h]))
            # bbox_dict["segmentation"] = [[xmin, ymin, xmax, ymin, xmax, ymax, xmin, ymax]]
            bbox_dict["segmentation"] = [list(np.round([xmin, ymin, xmax, ymin, xmax, ymax, xmin, ymax], 1))]
            kpts = line.strip().split(' ')[6:]
            for k_index, i in enumerate(kpts):
                if k_index % 3 == 2:
                    kpts[k_index] = 2
                else:
                    kpts[k_index] = int(float(i))
            bbox_dict["keypoints"] = kpts



            write_json_context["annotations"].append(bbox_dict)  # 将每一个由字典存储的bounding box 信息添加到'annotations'列表中

        name = os.path.join(json_path, "val" + '.json')
        # print(write_json_context)
        print(json.dumps(write_json_context))
        # with open(name,'w') as fw:    # 将字典信息写入.json文件中
        #     json.dump(write_json_context, fw, indent=2)
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

通过coco annotator导入后

注意：Datasets要设置category才会显示关键点信息