Real-Time AI Face Landmark Detection

Overview

The Real-Time AI Face Landmark Detection project demonstrates real-time facial landmark detection using TensorFlow.js and React. This project utilizes TensorFlow.js's FaceMesh model to identify and visualize facial landmarks in a live video stream from a webcam. The application is built using React and leverages the react-webcam library for video capture.

Features

• Real-Time Face Detection: Uses TensorFlow.js FaceMesh model to detect facial landmarks in real-time.
• Live Webcam Feed: Captures live video feed from the user's webcam.
• Facial Landmark Visualization: Draws facial landmarks and mesh on the canvas overlaying the webcam feed.
• Optimized for Performance: Real-time detection with efficient rendering for smooth user experience.

Technologies Used

• React: For building the user interface.
• TensorFlow.js: For face landmark detection with the FaceMesh model.
• react-webcam: For capturing live video feed from the user's webcam.
• JavaScript: For application logic and interactivity.

Installation

1. Clone the Repository:

   git clone https://github.com/your-username/RealTimeAIFaceLandmarkDetection.git

2. Navigate to the Project Directory:

   cd RealTimeAIFaceLandmarkDetection/facemesh

3. Install Dependencies:

    npm install

This will start the application on http://localhost:3000 by default.

Usage

To use the application:

1. Ensure you have a working webcam connected to your computer.
2. Open the application in your browser.
3. The application will request permission to access your webcam.
4. Once permission is granted, you will see the live video feed with facial landmarks and mesh drawn on it in real-time.

Screenshots & Video

faceDetectionVideo.mp4

Known Issues

• Browser Compatibility: The application is optimized for modern browsers with support for WebGL and TensorFlow.js.

• Performance: Real-time detection performance may vary based on the device and browser.

Contributing

Contributions are welcome! If you have suggestions, bug reports, or feature requests, please open an issue or submit a pull request.

Contact

For any questions or inquiries, you can reach out to:

• Email: tusharguptagps@gmail.com

• LinkedIn: Link

• Portfolio: Link

License

This project does not include a license. Feel free to use the code for personal or educational purposes. For any commercial use, please contact the project owner.

Thank you for checking out the Real-Time AI Face Landmark Detection project!