SJFit

Dumbbell Bicep Curls
Cable Tricep Push Down
Dumbbell Shoulder Press

What it does

SJFit.Ai was designed for San Jose State University Students and anybody who wants a fit lifestyle. In short, this application tracks the physical activity that the user does. We can track the angles of joints by implementing APIs such; as MediaPipe, OpenCV, and Artificial Intelligence, allowing us to count the number of repetitions the user has done thus far in their workout set.

How we built it

MediaPipe API, OpenCV API, Python, Juypter Lab for python

Challenges we ran into

Some of the challenges that we ran into was getting MediaPipe to run on M1 and M2 Apple silicon CPUs, as it has some issues with these types of CPU, which caused a lot of debugging for us, another major problem we ran into was getting the application to run accurate 100% of the time, it would not fully track our movements accurate 100% of the time.

Accomplishments that we're proud of

We are proud that we were able to overcome these challenges build our project to work well, and be accurate 95% of the time, which we plan to improve on with more time given.

What we learned

We learned a lot about git, and pushing. Along with that, we learned about; MedaPipe, how different hardware affects the software we run, Juypter Lab, Python, Debugging, Troubleshooting, etc.

What's next for SJFit

We would like to make this into a full-blown app for iOS, and Android, and a web application version, allowing users to use our application from anywhere from the comfort of their home to on the go.