Inspiration

Many of us want to get fit and strong, but learning how to work out with heavy weights properly often requires a friend or expert to guide one along the process. Thus, we set out to build an app that will guide you in a workout and provide feedback on what was done well and how to improve.

Media available: https://drive.google.com/open?id=14TswTCDE6KNzEhw3yiu6sxRBGTfA4l2W

What it does

WorkoutBud gives you real time guidance on your workout by displaying an animation overlay over your image. After a workout, the workout video can be analyzed to provide more details on how to improve form and give an overall score for workout effectiveness, so you can't cheat by doing bicep curls insanely fast!

How I built it

We used AlwaysAI Pose Estimator to do both the realtime and post-workout analysis. OpenCV was used to overlay an animation onto the live video.

Challenges I ran into

We ran into challenges regarding latency using OpenCV and AlwaysAI together. Since working out is very time sensitive, trying to display an animation on top of doing pose estimation was very challenging.

Accomplishments that I'm proud of

We are proud of the progress we made in creating a live feedback guidance animations, the data analysis for post-workout analysis, and the concept designs we had for the project idea.

What I learned

We learned a great deal about working with pose estimators and openCV to calculate animation, as well as how to analyze realtime data! We also learned how to do the perfect bicep curl.

What's next for WorkoutBud

WorkoutBut will see improvements with guidance latency and implementation of post-workout dashboard into iOS, Android, Desktop, and tablets.

Built With

Share this project:

Updates