Wealth is Health

Wealth is Health

Inspiration

Research shows that people who exercise regularly have better mental health, emotional well-being, and lower rates of mental illness. But according to the CDC, just 28% of Americans are exercising enough. To address this, we created an app that utilizes AI to accurately track how many repetitions of an exercise have been completed. We also included in-game currency, motivating users to exercise daily and reach their goals.

How we built it

We used MediaPipe's Pose Landmark Detection model to detect different segments of the body and used a math algorithm to determine if the user completed an exercise (pushup, situp, or squat). We used Tkinter to create the UI for the program and OpenCV to easily process the live webcam video.

Challenges we ran into

One challenge we ran into was the math and logic required in calculating whether the user completed an exercise. This used vector addition in the third dimension with the XYZ-Coordinate system that we learned in our math classes. The vector addition allowed us to match up the users' body position with what the desired exercise form looks like, to determine if the user completed an exercise. We also had problems with integrating MediaPipe's detection into the Tkinter GUI.

Accomplishments that we're proud of

We're proud of implementing an AI model and completing this project on time as this was our first time participating in a hackathon for 2 of the 3 members in our group.

What we learned

We learned how to use OpenCV, Tkinter, and MediaPipe's AI models, and how to integrate all of these tools together.

What's next for Wealth is Health

In the future, we want to improve our program by implementing more exercises that can be detected and include feedback to the user on their exercise form. A cloud saving and account feature can be added to the app to make it more convenient for users.