We felt that there was no crowd information on campus. We decided to come up with a solution to solve that problem.
What it does
Informs the user of the crowd density level at a particular study area.
How we built it
We used python scripts, opencv and pre-trained YOLOv2 Algorithm. This software was then made to run on a Raspberry Pi 4 with a Raspberry Pi Camera module.
Challenges we ran into
We had trouble setting up the raspberry pi without a monitor and getting the basic pre-trained model up on opencv.
Accomplishments that we are proud of
We somehow managed to pull through as a team :D
What we learned
Computer vision is not that straight forward, but it was fun to figure it out on such a short time notice.
What's next for NoHiding
We intend to use what we learn to apply CV techniques and AI techniques in our future projects.