MediGamer

Inspiration

As students, we often find ourselves getting up from our chairs after long study/gaming sessions with sore backs, wrists, and necks. Professional gamers like RNG Uzi face this issue severely, having to retire due to pain caused by extended sessions of sitting with bad posture. We all have seen companies tackle this issue, using ergonomic chairs, wrist rests, and pillows, but found these solutions ineffective, being uncomfortable, ineffective, or annoying, and not addressing the root of the issue: building up a habit of good posture.

What it does

Our project seeks to solve this issue, utilizing machine learning to recognize a user's worsening posture and alerting them with audio cues, without inturrupting intense gaming sessions. Our project also scores users based on their posture, with a "damage bar" that shows overall posture, and individual scores for back, neck, and wrist posture.

How we built it

Currently, our project runs on python using OpenCV and Google's Mediapipe Framework.

Challenges we ran into

Our biggest challenge was thinking of an interface where we could effectively alert users to fix their posture without inturrupting them at crucial moments of a game.

Accomplishments that we're proud of

We're proud of being able to put this togeter 100% virtually, collaborating through Discord.

What we learned

We learned how we can better mitigate negative health consequences of sitting for long durations of time.

What's next for MediGamer

We hope to be able to build this in Electron and Mediapipe JS in the future, giving a more polished experience, and also adding a feature to guide users through basic stretches they can do between games, or during loading sequences.