People Counter

What?

Our classroom was always busy and sometimes too loud to be able to work in a concentrated manner. We thought that having a second classroom might help resolve this problem. From this problem, the 'People Counter' was born. With this project, our goal was to count the number of people in the classroom and try to prove that we really needed a second classroom.

How?

I was tasked with making sure our model only detected people in the room and not all items that our model was capable of detecting. Ensuring that our running model was implemented as a subroutine of another script was also something I was working on. This was necessary because we wanted to be able to run the model on a Raspberry Pi.

Conclusion

The final result of this project was a camera setup in our classroom that can detect the number of people currently in the classroom and calculate a 10 minute average. Later this average would be posted to be shown on a discord bot. My first real touch with AI was through this project. Using YOLOv5, this was a fun way to get introduced to the different aspects of AI/ML/DL. I learned how to use the YOLOv5 model with python.