In recent times, a new social paradigm has emerged due to COVID-19. People are unable to exit their homes for long periods of time, and as a result their physical fitness has been going down considerably. Additionally, the lack of interaction with friends and family has been taking a mental and social toll on a lot of us as well.

As college students, we know first-hand what the lack of social and physical freedom can do. So we set out to solve this issue in a simple and easy-to-use manner. Y2K was the result of our work.

Y2K is a socially-oriented fitness app. Users can register on Y2K and find other users to exercise with using a simple web interface. The crux of the app is the dynamic inference of yoga poses. The user can exercise by holding a series of poses for a fixed amount of time [determined by the aforementioned ML model] and can also choose to exercise with a friend. The whole inference is all done in the browser to assuage privacy concerns. The track we used is DashaAI. The app was written in flask, mediapipe, HTML, CSS, JS, DashaAI. The USP of the app is this exercising with friends feature, as users are able to engage in a dynamic exercise with a friend and also interact with someone face to face at the same time. This web app also allows to release stress by meditation.

Built With

Share this project:

Updates