Skip to content

README.md

Latest commit

 

History

History
109 lines (65 loc) · 2.28 KB

README.md

File metadata and controls

109 lines (65 loc) · 2.28 KB

ASL Hand Sign Recognition

This project uses MediaPipe to extract hand landmarks and a Random Forest model to recognize American Sign Language (ASL) alphabet letters. It also includes real-time sign recognition using your webcam.

📌 Features

  • Uses the ASL Alphabet dataset from Kaggle
  • Extracts hand landmarks using MediaPipe
  • Trains a Random Forest classifier on landmark data
  • Tests accuracy on validation data
  • Predicts ASL hand signs in real-time via webcam

🛠 Requirements

Install the required Python libraries:

pip install mediapipe opencv-python scikit-learn joblib tqdm

🚀 How to Use

📥 Step 1: Download Dataset

  1. Download the ASL Alphabet Dataset from Kaggle:
    https://www.kaggle.com/datasets/grassknoted/asl-alphabet
  2. Extract it and place it inside a folder named dataset in the root of this project.

Step 2 : Extract the labels from Dataset

python data_extractor.py

Step 3: Train the Model

Run the training script:

python train.py

This will train and save the model as asl_model.pkl.

Step 4: Test the Model

Evaluate the model:

python test.py

Step 5: Real-Time Prediction

Start real-time webcam recognition:

python predict.py
  • Ensure your webcam is working.
  • Show your hand clearly in front of the camera.
  • Press q to quit.

⚙️ How It Works

  • MediaPipe detects 21 hand landmarks (x, y, z coordinates).
  • These landmarks are used as features.
  • A Random Forest classifier is trained on this feature set to predict ASL letters.
  • OpenCV manages real-time webcam access and display.

📝 Notes

  • The model uses landmark coordinates, not raw images, for better efficiency.
  • For best results: use a clean background and keep your hand visible and centered.
  • Supports static ASL letters (A–Z), excluding motion-based gestures like "J" and "Z".

🤝 Contributing

Pull requests are welcome! Feel free to fork or open an issue.

📄 License

This project is open-source under the MIT License.

👨‍💻 Author

Vivan Rajath
GitHub: https://github.com/VivanRajath