A real-time American Sign Language (ASL) letter detection system using OpenCV and MediaPipe. This project uses computer vision to detect and interpret hand gestures corresponding to ASL letters.
- Real-time hand tracking and landmark detection
- ASL letter recognition based on finger angles
- Visual feedback with hand landmarks
- Display of finger angles and normalized values
- Mirrored display for intuitive interaction
- Web interface for easy access
- Support for all 26 ASL alphabet letters
- Python 3.7+
- Webcam or camera device
- Clone the repository:
git clone https://github.com/yourusername/OpenCV-ASL-Reader.git
cd OpenCV-ASL-Reader- Install the required dependencies:
pip install -r requirements.txtThere are two ways to run the ASL Reader:
- Start the web server:
python app.py- Open your web browser and navigate to:
http://localhost:5000
- The webcam feed will start automatically in your browser
- Run the standalone application:
python asl_reader.pyIn both versions, the application will:
- Display hand landmarks
- Show finger angles and normalized values
- Predict and display ASL letters in real-time
- Press 'q' to quit (terminal version only)
- Supports all 26 letters of the ASL alphabet
- Real-time detection and visualization
- User-friendly web interface
- Detailed angle measurements and normalization
- Mirrored display for intuitive interaction
- Implement machine learning for better accuracy
- Add support for dynamic gestures
- Improve robustness in various lighting conditions
- Add word and sentence recognition
- Add tutorial mode for learning ASL
Contributions are welcome! Please feel free to submit a Pull Request.
This project is licensed under the MIT License - see the LICENSE file for details.
- Enable GitHub Pages in your repository settings
- Set the source to the main branch
- Your frontend will be available at
https://[username].github.io/OpenCV-ASL-Reader
- Create a new Heroku app
- Add Python buildpack
- Deploy the backend: