American Sign Language Translator

This project is a real-time American Sign Language (ASL) translator that uses a webcam to detect and classify hand gestures into corresponding letters of the alphabet. The system utilizes computer vision and machine learning to identify hand landmarks and make predictions based on a trained model.

Table of Contents

• Demo
• Features
• Installation
• Usage
• Project Structure
• Training the Model
• Real-Time Prediction
• Contributing
• License

Demo

Click on the image to watch the demo video.

Features

• Real-time Hand Gesture Recognition: Uses a webcam to capture hand movements and predict corresponding ASL letters.
• Machine Learning Model: Utilizes a Random Forest classifier trained on hand landmarks data.
• Simple and User-Friendly: Easy to set up and use for educational or demonstration purposes.

Installation

1. Clone the Repository:

   git clone https://github.com/your-username/ASL-Translator.git
   cd ASL-Translator

2. Install Dependencies:

   Install the required packages using pip:

   pip install -r requirements.txt

3. Additional Requirements:

   • Ensure you have a working webcam.
   • Python 3.10 or higher is recommended.

Usage

1. Run the Real-Time ASL Translator:

   Navigate to the project directory and run the realtime_predictor.py script:

   python realtime_predictor.py

2. Instructions:

   • Make sure your webcam is connected.
   • The program will start capturing video. Show a hand gesture in front of the camera.
   • The corresponding ASL letter will be displayed on the screen.
   • Press E to exit the program.

Project Structure

.
├── data/                          # Directory containing image data for training
├── model/                         # Directory for saving trained models
│   └── model.p                    # Pre-trained model file
├── src/
│   ├── dataset_creator.py         # Script for creating and saving the dataset
│   ├── model_trainer.py           # Script for training the model
│   └── realtime_predictor.py      # Script for real-time prediction
├── requirements.txt               # Project dependencies
└── README.md                      # Project documentation

Training the Model

To train the model on your own dataset:

1. Organize Data:

   Place your hand gesture images in the data directory, organized into subdirectories where each subdirectory name corresponds to the class label (e.g., 0, 1, 2, ..., 25).

2. Generate Dataset:

   Run the dataset_creator.py script to create the dataset:

   python src/dataset_creator.py

3. Train the Model:

   Run the model_trainer.py script to train the model:

   python src/model_trainer.py

4. The trained model will be saved in the model directory as model.p.

Real-Time Prediction

To use the real-time prediction functionality:

1. Ensure your webcam is connected.

2. Run the realtime_predictor.py script:

   python src/realtime_predictor.py

3. The system will start capturing video, detecting hand gestures, and displaying predictions in real-time.

Contributing

Contributions are welcome! If you'd like to contribute, please fork the repository and create a pull request with your changes. For major changes, please open an issue first to discuss what you would like to change.

License

This project is licensed under the MIT License - see the LICENSE file for details