Sign Language Recognition

Sign Language Recognition is a project for recognizing sign languages used by hearing impaired people.

Project Features

1. Real-Time Hand Gesture Recognition Model.
2. Detects sign language based hand gestures used by deaf people
3. As of now detects English Alphabet (A-Z) signs.
4. Trained on multiple kind of hand images viz. different skin colours, different sizes.
5. Dataset was created using OpenCv by clicking images and also storing them in the named

Usage

The idea behind the project is to lay a foundation for a major project which will help hearing impaired people learn the sign language using recognition, use the same to get employment and in various industries such as Vending machines, displays in stores etc.

Software Requirement

1. Python 3.5–3.7
2. Tensorflow
3. pip 19.0 or later (requires manylinux2010 support)
4. Ubuntu 16.04 or later (64-bit) or macOS 10.12.6 (Sierra) or later (64-bit) (no GPU support) or Windows 7 or later (64-bit) (Python 3 only)
5. Microsoft Visual C++ Redistributable for Visual Studio 2015, 2017 and 2019
6. Raspbian 9.0 or later
7. GPU support requires a CUDA®-enabled card (Ubuntu and Windows)
8. VS Code/Anaconda

Images type and Snapshots of Working Model

1. Signs for Hearing Impaired

2. Snapshot of a part of Dataset

3. Real time sign recognition of letter B

4. Real time sign recognition of letter D

References

1. Stack Overflow : www.stackoverflow.com
2. Google : www.google.co.in
3. Tensorflow : www.tensorflow.org
4. Youtube : www.youtube.com
5. Nicholas Renotte : www.nicholasrenotte.com

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

Please make sure to update tests as appropriate.