✋ SignGT

SignGT is an AI-powered sign language recognition and translation system that uses computer vision and machine learning techniques to detect hand gestures and convert them into meaningful text in real time. The project aims to bridge the communication gap between people who use sign language and those who do not.

🚀 Features

• Real-time sign language detection
• Gesture recognition using computer vision
• Translation of signs into text
• User-friendly interface
• Extensible architecture for adding new signs and languages
• Efficient preprocessing and prediction pipeline

🛠️ Tech Stack

• Programming Language: Python
• Machine Learning: TensorFlow / PyTorch
• Computer Vision: OpenCV, MediaPipe
• Frontend: Streamlit / Flask (if applicable)
• Data Processing: NumPy, Pandas
• Visualization: Matplotlib

📂 Project Structure

SignGT/
├── dataset/              # Training and testing datasets
├── models/               # Trained model files
├── notebooks/            # Jupyter notebooks for experiments
├── src/
│   ├── preprocessing.py
│   ├── train.py
│   ├── predict.py
│   └── utils.py
├── app.py                # Main application
├── requirements.txt
├── README.md
└── LICENSE

⚙️ Installation

1. Clone the repository

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

2. Create a virtual environment

python -m venv venv

Activate it:

Windows

venv\Scripts\activate

Linux/macOS

source venv/bin/activate

3. Install dependencies

pip install -r requirements.txt

▶️ Running the Project

python app.py

If using Streamlit:

streamlit run app.py

🧠 Model Pipeline

1. Capture hand gestures using a webcam.
2. Extract hand landmarks using MediaPipe/OpenCV.
3. Preprocess the extracted features.
4. Feed the features into the trained model.
5. Convert predictions into readable text.

📊 Dataset

The model is trained on a dataset containing images/videos of sign language gestures.

Dataset includes:

• Hand gesture images
• Landmark coordinates
• Corresponding labels

Replace this section with details about your dataset source and size.

📈 Results

Metric	Value
Accuracy	74%
Precision	68%

🎯 Future Improvements

• Support for complete sentence generation
• Multi-language sign recognition
• Speech synthesis from predicted text
• Mobile application deployment
• Improved model accuracy with larger datasets

🤝 Contributing

Contributions are welcome!

1. Fork the repository
2. Create a new branch
3. Commit your changes
4. Push to your branch
5. Open a Pull Request

📜 License

This project is licensed under the MIT License.

👩‍💻 Author

Kamakshi Mudgal