深度学习项目十七：YOLOv8关键点pose训练自己的数据集

* 2.2 格式转换（非常重要）
* 2.3 数据集划分

    pip install labelme
    
    
      
    
    代码解读

    # 将labelme标注的json文件转为yolo格式
    import os
    import cv2
    import numpy as np
    import matplotlib.pyplot as plt
    import glob
    import json
    import tqdm
    
    
    # 物体类别
    class_list = ["hongbi", "luosi", "qianbi", "makebi"]
    # 关键点的顺序
    keypoint_list = ["head", "tail"]
    
    
    def json_to_yolo(img_data,json_data):
    h,w = img_data.shape[:2]
    # 步骤：
    # 1. 找出所有的矩形，记录下矩形的坐标，以及对应group_id
    # 2. 遍历所有的head和tail，记下点的坐标，以及对应group_id，加入到对应的矩形中
    # 3. 转为yolo格式
    
    rectangles = {}
    # 遍历初始化
    for shape in json_data["shapes"]:
        label = shape["label"] # pen, head, tail
        group_id = shape["group_id"] # 0, 1, 2, ...
        points = shape["points"] # x,y coordinates
        shape_type = shape["shape_type"]
    
        # 只处理矩形
        if shape_type == "rectangle":
            if group_id not in rectangles:
                rectangles[group_id] = {
                    "label": label,
                    "rect": points[0] + points[1],  # Rectangle [x1, y1, x2, y2]
                    "keypoints_list": []
                }
    # 遍历更新，将点加入对应group_id的矩形中
    for keypoint in keypoint_list:
        for shape in json_data["shapes"]:
            label = shape["label"]
            group_id = shape["group_id"]
            points = shape["points"]
            # 如果匹配到了对应的keypoint
            if label == keypoint:
                rectangles[group_id]["keypoints_list"].append(points[0])
    
    # 转为yolo格式
    yolo_list = []
    for id, rectangle in rectangles.items():
        result_list  = []
        label_id = class_list.index(rectangle["label"])
        # x1,y1,x2,y2
        x1,y1,x2,y2 = rectangle["rect"]
        # center_x, center_y, width, height
        center_x = (x1+x2)/2
        center_y = (y1+y2)/2
        width = abs(x1-x2)
        height = abs(y1-y2)
        # normalize
        center_x /= w
        center_y /= h
        width /= w
        height /= h
    
        # 保留6位小数
        center_x = round(center_x, 6)
        center_y = round(center_y, 6)
        width = round(width, 6)
        height = round(height, 6)
    
    
        # 添加 label_id, center_x, center_y, width, height
        result_list = [label_id, center_x, center_y, width, height]
    
        # 添加 p1_x, p1_y, p1_v, p2_x, p2_y, p2_v
        for point in rectangle["keypoints_list"]:
            x,y = point
            x,y = int(x), int(y)
            # normalize
            x /= w
            y /= h
            # 保留6位小数
            x = round(x, 6)
            y = round(y, 6)
            
            result_list.extend([x,y,2])
    
        yolo_list.append(result_list)
    
    return yolo_list
    
    
    # 获取所有的图片
    img_list = glob.glob("./auto_labeled/*.jpg")
    
    for img_path in tqdm.tqdm( img_list ):
        
    img = cv2.imread(img_path)
    
    json_file = img_path.replace('jpg', 'json')
    with open(json_file) as json_file:
        json_data = json.load(json_file)
    
    yolo_list = json_to_yolo(img, json_data)
    
    yolo_txt_path = img_path.replace('jpg', 'txt')
    with open(yolo_txt_path, "w") as f:
        for yolo in yolo_list:
            for i in range(len(yolo)):
                if i == 0:
                    f.write(str(yolo[i]))
                else:
                    f.write(" " + str(yolo[i]))
            f.write("\n")
    
    
    
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
      
    
    代码解读

    from ultralytics import YOLO
    # Load a model
    model = YOLO("/home/xiao/ultralytics-main/ultralytics/cfg/models/v8/yolov8-pose.yaml")
    model = YOLO("/home/xiao/ultralytics-main/weights/yolov8s-pose.pt")
    model = YOLO("/home/xiao/ultralytics-main/ultralytics/cfg/models/v8/yolov8-pose.yaml").load("/home/xiao/ultralytics-main/weights/yolov8s-pose.pt")
    
    # Train the model
    results = model.train(data="/home/xiao/ultralytics-main/ultralytics/cfg/datasets/fks-pose.yaml", epochs=300, imgsz=1280, batch=2)
    
    
    
      
      
      
      
      
      
      
      
      
    
    代码解读

    train: /home/xiao/dataset/fks14-24/images/train
    val: /home/xiao/dataset/fks14-24/images/val
    
    # Keypoints
    kpt_shape: [4, 3] # number of keypoints, number of dims (2 for x,y or 3 for x,y,visible)
    # flip_idx: [0, 1, 2, 3]
    
    # Classes
    names:
      0: hongbi
      1: luosi
      2: qianbi
      3: makebi
    
    
      
      
      
      
      
      
      
      
      
      
      
      
      
    
    代码解读

    kpt_shape: [2, 3]  # number of keypoints, number of dims (2 for x,y or 3 for x,y,visible)
    
    
      
    
    代码解读

    from ultralytics import YOLO
    
    
    # Load a model
    model = YOLO("/home/xiao/ultralytics-main/weights/yolov8s-pose.pt")                # load an official model
    model = YOLO("/home/xiao/ultralytics-main/runs/pose/train6/weights/best.pt")       # load a custom model
    
    
    # Predict with the model
    results = model("/home/xiao/ultralytics-main/6-25test/", save=True)                  # predict on an image
    
    
      
      
      
      
      
      
      
      
      
      
    
    代码解读