FacePoseFusion integrates real-time face detection and pose estimation using MediaPipe. It captures video input, processes each frame to overlay facial and body landmarks, and saves the output as an MP4 file. Ideal for applications in gaming and AR, it showcases advanced machine learning in practical use.
FacePoseFusion is a real-time application that integrates face detection and pose estimation using MediaPipe, an open-source framework by Google. This project captures video input, processes each frame to detect facial features and human body landmarks, and overlays the results on the live video feed.
- Real-Time Face Detection: Identifies and tracks faces in the video feed with high accuracy.
- Human Pose Estimation: Estimates body poses by detecting key landmarks, enabling interactive applications.
- Dynamic Video Processing: Calculates frames per second (FPS) dynamically for smooth video playback.
- Output Video Saving: Saves the processed video as an MP4 file for further analysis or sharing.
- Python
- OpenCV
- MediaPipe
- Clone the repository:
git clone https://github.com/yourusername/FacePoseFusion.git cd FacePoseFusion