Skip to content

hackslash-nitp/signlanguagedetection

Repository files navigation

signlanguagedetection

Deployment URL :-

https://sign-lang-website.herokuapp.com/

Demonstration Video :-

https://drive.google.com/file/d/1bPppyCxwb-HcGY-INP8XVlTMAAn30_aL/view?usp=sharing

✨The key points making this one exceptional are that the deep CNN model has been trained on over 500 real-time high-quality images captured from the webcam and is capable to detect hand signs from 0-9 (which can easily be extended to include A-Z).

✨The robustness of this solution lies in the fact that its training has been done against different environment weights and runs on Web GL without any issues of platform resources or downtime.

👉 Sign Language standard used for model training :-

Image of Model

Team Gray-Interface ✖️ HackSlash

Mentors - Divya Prakash, Suraj kumar, and PARITOSH.