A language learning tool for American Sign Language which uses skeletal tracking data from a Leap Motion and machine learning to interpret signs.
Please watch the video demo!
This project was made in under 24 hours for TAMUHack 2015. Thanks to the organizers!
- Clone this repository
- Install packages from
requirements.txt
- Start Redis running on localhost at port 6379
python app.py
- This step may take a few seconds the first time because the classifier has to be trained on the data. After that first time, the classifier is serialized to disk and loaded much more quickly on subsequent restart.
- Open a browser to the URL outputted in the console