Inspiration
As a teaching assistant for a class, one member in our team wanted to create a scalable IoT and ML-based biometric attendance monitoring system for classes. We thought this would help professors and TAs not waste valuable teaching time.
What it does
Given a photo roster for a class, it models the face of each student. Then we place a microcontroller with attached camera near the door of a lecture hall and it tracks people as the leave and enter and tracks the time they spent in the class. This sends the data to a RESTful server and the students are marked present or absent depending on the number of minutes they were in the lecture. It can also detect multiple people entering or leaving at the same time.
How we built it
We use Python and OpenCV to do the facial recognition and Flask for the RESTful backend server.
Challenges we ran into
It was very hard to allow multiple people to be incorporated into our facial recognition model. We also had a lot of version control issues as all of us were trying to work on the same files.
Accomplishments that we're proud of
We were able to create a robust attendance system in less than 24 hours, which will make classrooms much better.
What we learned
How to do better version control!
What's next for ArKAL
Implementing it for non-classroom purposes
Log in or sign up for Devpost to join the conversation.