菜单
Implement real-time object detection using YOLOv5 constrained to a specific Region of Interest (ROI) polygon, utilizing GPU acceleration and the supervision library for annotation.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/ECNU-ICALK/AutoSkill --skill yolov5-object-detection-with-roi-masking-and-gpu-support命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| id | ee4e0037-ac33-4813-aad9-0a950ddcc6c2 |
| name | YOLOv5 Object Detection with ROI Masking and GPU Support |
| description | Implement real-time object detection using YOLOv5 constrained to a specific Region of Interest (ROI) polygon, utilizing GPU acceleration and the supervision library for annotation. |
| version | 0.1.0 |
| tags | ["yolo","computer-vision","roi","gpu","python","opencv"] |
| triggers | ["yolov5 roi masking","detect objects in polygon area","yolov5 gpu inference","supervision library yolo"] |
Implement real-time object detection using YOLOv5 constrained to a specific Region of Interest (ROI) polygon, utilizing GPU acceleration and the supervision library for annotation.
Act as a Computer Vision Engineer. Write Python code to perform real-time object detection using YOLOv5, constrained to a specific Region of Interest (ROI) defined by a polygon. The code must run on GPU if available.
torch.hub.load('ultralytics/yolov5', 'yolov5s6', device=device).device = 'cuda' if torch.cuda.is_available() else 'cpu'.np.array([[x1,y1], [x2,y2], ...], dtype=np.int32)).cv2.bitwise_and to mask the frame.supervision.BoxAnnotator to draw boxes on the original (unmasked) frame.Provide the complete, runnable Python script including imports and the main execution loop.