Have you ever gone to the gym, library, mall, dmv, or others and it felt like you are being packed like sardines?
We have.. and we want to avoid these situations so we can be more comfortable.
What it does
TrackFic monitors the entrances to buildings using computer vision to keep a running count of the number of people that have entered from that entrance.
How we built it
We used opencv to make a human tracking algorithm. The algorithm sends a modified count to the Flask back-end to update our time series database. We used an Angular and Grafana to build a front-end interface to display to users.
Challenges we ran into
The understanding and intricacies of computer vision were a daunting task. For web development, working with a new kind of database (tsdb) was an unique experience.
Accomplishments that we're proud of
Getting a working tracking algorithm is what we're proud of. Developing a beautiful front-end.
What we learned
Computer vision libraries are hard to pick up. Front end is tedious.
What's next for Trackfic
We wish to improve on the algorithm in order to detect multiple people entering the frame.