yolov8Pose实战

前言

YOLO系列层出不穷，从yolov5到现在的yolov8仅仅不到一年的时间。追踪新技术，了解前沿算法，一起来测试下yolov8在不同人物如pose检测的效果吧！

一、yolov8环境搭建

1.1 Source code: git

1.2 安装必要的package：

安装须知：yolo5是可以兼容python3.7以及对应的numpy; 而yolo8使用python3.8以上。
不然会遇到问题：

1. TypeError: concatenate() got an unexpected keyword argument ‘dtype’ #2029

• 解决办法：conda一个新的环境，python=3.8 并直接pip install ultralytics

1. 按照上述操作，还遇到bug：
   F.conv2d(input, weight, bias, self.stride,
   RuntimeError: GET was unable to find an engine to execute this computation"

• 解决方法参考：Issue

此外，国内防火墙有可能会禁止某些package的下载，特别是与pytorch相关的大文件。可参考国内镜像下载包

二、测试

训练模型，评估模型，并导出模型

from ultralytics import YOLO

# Create a new YOLO model from scratch
model = YOLO('yolov8n.yaml')

# Load a pretrained YOLO model (recommended for training)
model = YOLO('yolov8n.pt')

# Train the model using the 'coco128.yaml' dataset for 3 epochs
results = model.train(data='coco128.yaml', epochs=3)

# Evaluate the model's performance on the validation set
results = model.val()

# Perform object detection on an image using the model
results = model('https://ultralytics.com/images/bus.jpg')

# Export the model to ONNX format
success = model.export(format='onnx')

可以成功训练评估以及导出onnx模型文件．

实测检测效果

拿一张照片去测试上述训练后的模型

from ultralytics import YOLO
from PIL import Image
import cv2

model = YOLO("model.pt")

# accepts all formats - image/dir/Path/URL/video/PIL/ndarray. 0 for webcam
results = model.predict(source="0")
results = model.predict(source="folder", show=True) # Display preds. Accepts all YOLO predict arguments

# from PIL
im1 = Image.open("bus.jpg")
results = model.predict(source=im1, save=True)  # save plotted images

# from ndarray
im2 = cv2.imread("bus.jpg")
results = model.predict(source=im2, save=True, save_txt=True)  # save predictions as labels

# from list of PIL/ndarray
results = model.predict(source=[im1, im2])

实测效果如下：

测试人体姿态估计

打开摄像头，检测人物的姿态

from ultralytics import YOLO
import cv2
import math
import os
import glob
import numpy as np

# Load a model
model = YOLO('yolov8n-pose.pt')  # load an official model
# model = YOLO('path/to/best.pt')  # load a custom trained

video_path = 0
cap = cv2.VideoCapture(video_path)

while cap.isOpened():
    success, frame = cap.read()
    results = model(frame, imgsz=256)
    annotated_frame = results[0].plot()
    print(results[0].tojson('data.json'))
    cv2.imshow("YOLOv8 pose inference", annotated_frame)
    if cv2.waitKey(1) & 0xFF == ord("q"):
        break

cap.release()
cv2.destroyAllWindows()

这是打开本地摄像头实测的效果图，实时性能佳