Inspiration

We wanted to make personal fitness feel more accessible and help to build a healthy community.

What it does

SpotBot does real-time pose tracking to log workout repetitions and sets.

How we built it

We used MediaPipe, React, Typescript, React Native, Android.

Challenges we ran into

It took us 6 hours to get our tech-stack together.

Accomplishments that we're proud of

We're proud that we were able to build something that we were interested in and work together to accomplish something that we thought was out of our wheelhouse.

What we learned

We learned about state-of-the-art computer vision technology, as well as integrating third-party services into React Native. We also learned how to develop mobile applications for Android.

What's next for SpotBot

We would like to integrate computer vision alternatives, as the current build uses MediaPipe for body landmark detection. However, other computer vision libraries and models like TensorFlow’s MOVENET could be useful, and there may be even faster alternatives. We would also like to include more complex movements for an expanded workout library, and save workout data for personalized & adaptive goals.

Share this project:

Updates