Inspiration
We often found ourselves struggling to find an open space to study with our group of friends. We have never found any solution at UMB where we could easily track what spaces were available to study or how crowded they were, so we wanted to fix that.
What it does
Our app uses cameras positioned at the entrance of a given study space to get an estimate of how crowded the space is. This data is saved in a SQL Lite database, and is then presented on a react frontend.
How we built it
We built it using OpenCV with a Django server for image and video processing and API requests, and React for frontend functionality.
Challenges we ran into
We had several challenges with video processing. Getting OpenCV to detect one person as, well one person and not a hundred little people put together was quite the challenge.
Accomplishments that we're proud of
Getting the OpenCV video detection was our biggest Accomplishment on the backend, and for the project as a whole. It took a lot of tweaking with object tracking algorithms we used to get it just right. As for the frontend, the use of context providers made sharing data across components was a big accomplishment as well.
What we learned
We learned video processing is extremely difficult, but we loved the challenge. We enjoyed the challenge it presented and we would love to improve on it in the future, because it isn't perfect
What's next for Study Spaces
The next step is fleshing out the frontend to be more sleek and robust. We also want to look into using Machine Learning for tracking people on a video feed rather than relying on static algorithms.
Log in or sign up for Devpost to join the conversation.