Inspiration
The project was inspired by one of the team member's family story. It involves inclusivity and fairness in the tech industry, as well as more representation for deaf peiple.
What it does
Our project converts ASL to text. Currently, it can translate the alphabet.
How we built it
We started our code using mediapipe, opencv, numpy, os, as well as online resources and books. After we ran into some design challenges, we had to switch some things to Ubuntu and Google Colab.
Challenges we ran into
When we first tried to train our AI model, we ran into library issues with the mediapipe model maker. However, we circumvented this by using a ubuntu vmware to compile our data, and then bring the compiled info back to our source code.
Accomplishments that we're proud of
We're proud of finishing a complex project in less than a day and how no one had to deal with a single challenge independently. We worked well in a team, despite the intensity and pressure of a codensed timeline.
What we learned
Our team learned a lot of technical skills such as, mediapipe, opencv, ubuntu, and pycharm, however, most importantly, we learned how to work in a team. It's hard to just jump into the project without a plan, so we divided up the work, but we didn't let anyone work alone.
What's next for Gesture Recognition
Refine and fine-tune the project so we can incorporate more features such as word structure and speech.
Log in or sign up for Devpost to join the conversation.