yolov8使用opencv 调用摄像头，实时目标检测、分割、关键点检测，左上角显示帧率

地里的小趴菜

已于 2024-07-20 09:06:25 修改

阅读量554

点赞数 2

文章标签： YOLO opencv 目标检测

于 2024-07-19 19:19:16 首次发布

本文链接：https://blog.csdn.net/lbr15660656263/article/details/140557854

版权

一、目标检测

import cv2
from ultralytics import YOLO

# 加载 YOLOv8 模型
model = YOLO('yolov8n.pt')
video_path =0 #摄像头ip地址
cap = cv2.VideoCapture(video_path)
while cap.isOpened():
    status, frame = cap.read()
    if not status:
        break
    results = model.predict(source=frame)
    result = results[0]
    anno_frame = result.plot()
    cv2.imshow("V8", anno_frame)
    # 设置延迟以减慢视频播放速度
    if cv2.waitKey(30) & 0xFF == ord('q'):
        break
cap.release()
cv2.destroyAllWindows()

二、关键点检测

import cv2
from ultralytics import YOLO

# 加载 YOLOv8 模型
model = YOLO('yolov8n-pose.pt')
video_path =0 #摄像头ip地址
cap = cv2.VideoCapture(video_path)
while cap.isOpened():
    status, frame = cap.read()
    if not status:
        break
    results = model.predict(source=frame)
    result = results[0]
    anno_frame = result.plot()
    cv2.imshow("V8", anno_frame)
    # 设置延迟以减慢视频播放速度
    if cv2.waitKey(30) & 0xFF == ord('q'):
        break
cap.release()
cv2.destroyAllWindows()

三、分割

import cv2
from ultralytics import YOLO

# 加载 YOLOv8 模型
model = YOLO('yolov8n-seg.pt')
video_path =0 #摄像头ip地址
cap = cv2.VideoCapture(video_path)
while cap.isOpened():
    status, frame = cap.read()
    if not status:
        break
    results = model.predict(source=frame)
    result = results[0]
    anno_frame = result.plot()
    cv2.imshow("V8", anno_frame)
    # 设置延迟以减慢视频播放速度
    if cv2.waitKey(30) & 0xFF == ord('q'):
        break
cap.release()
cv2.destroyAllWindows()