Video Broadcaster is a real-time video streaming application that utilizes AI-powered segmentation to apply various background effects. This project is designed to work with virtual camera software, allowing users to broadcast video with customized backgrounds.
- Real-time Person Segmentation: Uses YOLOv8 for accurate person detection and segmentation.
- Background Effects: Choose from custom image backgrounds, blurred backgrounds, or black backgrounds.
- Adjustable Settings: Modify FPS and blur strength to suit your streaming needs.
- Virtual Camera Output: Stream processed video to virtual camera software for broadcasting.
- Web-based Control Interface: Manage streaming settings through a user-friendly web interface.
- Python: Version 3.8 or higher.
- Virtual Camera Software: OBS Virtual Camera or similar.
- CUDA-capable GPU: Recommended for optimal performance with YOLOv8.
- Clone the Repository:
git clone https://github.com/yourusername/video_broadcaster.git cd video_broadcaster pip install requirements.txt python main.py
- GET /api/settings: Get current streaming settings.
- GET /devices: Update streaming settings.
- GET /start: Starts streaming with params
- GET /stop: Stops streaming
This README provides comprehensive information about your project, including setup, usage, and troubleshooting. Let me know if you need any additional sections or modifications!