Sign in page
Home page
Activity log page

Inspiration

Every athlete knows someone whose season or career was cut short by injury. The frustrating part is that many injuries are not random—they are linked to modifiable movement patterns. But quality form coaching is expensive and inaccessible. We were inspired by the idea of giving every athlete, especially underprivileged kids who cannot afford a health coach, a free 24/7 way to check their form. If we can reduce even a small amount of risk per session, it compounds:

What it does

FormFit is a website where users upload a video of themselves playing a sport or doing an exercise—basketball shots, squats, etc. The AI analyzes the clip using biometric/pose tracking, computes joint angles, alignment, tempo, and symmetry, and returns simple feedback to improve form and reduce injury risk. It acts like a free health coach for anyone with a phone.

How we built it

We built FormFit as a full-stack web app. The frontend handles video upload and displays annotated feedback. The backend processes frames, runs pose estimation to extract keypoints, smooths noisy estimates, and computes biomechanical features. We use sport-specific rule sets for basketball and squats, with a hybrid rules + ML scoring approach. The AI uses pose/biometric tracking to analyze user videos and generate feedback on form. It is designed as a free, accessible health coach, targeting underprivileged kids who cannot afford personal coaching.

Challenges we ran into

The main challenge we ran into came from the idea and collaboration aspect. Coming up with an idea that we were genuinely passionate about as a team, as well as trying to come up with a balance of all the ideas we had created major roadblocks.

Accomplishments that we're proud of

We are proud of our end-to-end prototype, and the fact that we were able to create a final project that targeted athletes and underprivileged youth. We turned form feedback accessible 24/7 and free.

What we learned

We learned that time management is extremely important in projects like these as it took us a lot of time to think of our idea and considering we had to balance this with school work as well, the time we were left with to actually create our idea was quite minimal so we had to make the most of our time.

What's next for FormFit

Creating different modes where the AI can feedback on the athletes’ forms based on things besides injury prevention. For example, the AI could feedback on squat reps to teach the user how to improve their form in a way that increases muscle contraction and therefore increases strength gain.