ASL Hand Sign Recognition

A Streamlit-based ASL hand sign recognition app that uses MediaPipe and an ensemble of TensorFlow models for real-time webcam and image input.

Overview

This repository contains an ASL alphabet recognition system built on MediaPipe hand detection, OpenCV preprocessing, and a multi-model TensorFlow ensemble for reliable real-time inference.

Quick Start

python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
streamlit run app.py

Features

• Real-time ASL alphabet recognition using webcam input
• MediaPipe hand tracking with robust fallback handling
• Ensemble inference with multiple TensorFlow models
• Image upload support and confidence scoring
• Designed for accuracy and usability with a lightweight UI

Requirements

• Python 3.11+ recommended
• requirements.txt includes:
  • streamlit
  • streamlit-webrtc
  • tensorflow-cpu
  • opencv-python-headless
  • mediapipe
  • av
  • Pillow
  • numpy
  • huggingface-hub

Usage

1. Create and activate a Python virtual environment.
2. Install dependencies from requirements.txt.
3. Launch the app with streamlit run app.py.
4. Open the local Streamlit URL displayed in the terminal.

Repository layout

• app.py - main Streamlit application entrypoint
• requirements.txt - core Python dependencies
• packages.txt - duplicate dependency list included for compatibility
• class_names.json - ASL alphabet label definitions
• models/ - model artifacts and helper files
• all-train-data/ - training dataset notes and notebooks
• cheeklist-Ahmed_abobakr/ - project checklists and documentation
• digrams-Ahmed_abobakr/ - architecture and process diagrams
• improvmint-ahmed_sabary/ - improved ASL engine implementation and support files

Contributing

See CONTRIBUTING.md for issue and pull request guidelines.

License

This project is licensed under the MIT License. See LICENSE for details.

Policies

• Contributor behavior is defined in CODE_OF_CONDUCT.md.
• Security reporting is described in SECURITY.md.