Inspiration
A lot of kids find themselves disengaged and/or sleeping in class. Our team more definitely fall victim to this. According to a 2017 Gallup study, over 56% of high school students are disengaged and this is excluding those who've dropped out.
What it does
CodeCall scans the room for hands raised and using facial recognition, will present the teacher with a name that they can call on. If no hands are present it will randomly call on students who don't regularly participate.
How we built it
We used a flask framework to allow all of us to work together at once. As Jason figured out the apis and AI components, Jasmine worked on the frontend, while Benny and Calvin worked on integration. We leveraged mediapipes, a simple to use but powerful and lightweight google AI module for hand and face detection, then we used the Alchera Face Authentication API hosted on Rapid API in order to build the facial recognition.
Challenges we ran into
Unfortunately we ran into issues integrating these components and building our app further.
Accomplishments that we're proud of
Getting 2 out of the 3 AIs working was an accomplishment we're proud of.
What we learned
We learned how to use CV2, utilize AIs (TensorFlow), planning our development, and leveraging APIs to speed up development.
What's next for DuckDuckCall
We need to fix a crucial bug that is blocking integration between our different micro projects.
Log in or sign up for Devpost to join the conversation.