SignPrism

Opening or Splash screen of the app
Point the camera to a person's hand and the Tf model will start making predictions. Using this predicted letters we can form sentences.
An info page where all the letters with their respective hand signs will be shown.

Inspiration

We wanted to tackle the lack of communication one can face while communicating with deaf people..

What it does

Sign Prism is gateway to communication with deaf people. People who don't know any sign language can use this app to communicate with deaf people.

How we built it

We used Flutter to build the UI of the app and used dart, python for backend.

Challenges we ran into

While building this project we came we faced couple of challenges:

Initially the accuracy of the model was pretty low, so we had to tweak quite a lot of parameters to reach an acceptable level of accuracy.
We faced trouble while capturing hand signs in real time and simultaneously display those American Sign Language characters.

Accomplishments that we're proud of

We were able to build something that can eventually help people with disabilities and at the same time expand our technical knowledge

What we learned

While building this project we came across the difficulties deaf people face and how hard it is for them to communicate with someone who doesn't sign language.

What's next for SignPrism

We have couple of plans for SignPrism. In future we will try to increase the accuracy of our ML model and build a good looking UI. Integrating it into a video call so that it will become much more easier to communicate with deaf people.