Ultralytics (CPU) Object Detection

Ultralytics's YOLO CLI allows convenient execution of object detection on images and videos.

⚙️ Usage

First, open the project in GitHub Codespaces by clicking here. Then, follow the steps below.

Note: GitHub Codespaces are free for a limited time (180 hours). When not in use, an active Codespace should be stopped or deleted. You can manage your GitHub Codespaces by clicking here.

Steps

1. Downloading a Video

Use yt-dlp to retrieve a video from YouTube or a supported source. In the example, a video named cars.mp4 is downloaded:

yt-dlp -o cars.mp4 https://youtu.be/CftLBPI1Ga4

Note: Alternatively, a local video can be uploaded via Drag & Drop.

2. Running YOLO Object Detection

Use yolo to perform object detection on the video cars.mp4:

yolo predict model=yolov8n-seg.pt source='cars.mp4'

Note: The annotated video file is located in the /runs/predict/predict{n}/ folder.

Options:

• model: Specifies the YOLO model for object detection. For a list of available models, see here.
• source: Specifies the video source, which can be an image or video file.

👷‍ Error Found?

Thank you for your message! Please fill out a bug report.

License

This project is licensed under the European Union Public License 1.2.