-
test success!
Inspiration
Our friend hurt himself in the gym due to poor form and it resulted in months of recovery and chronic pain. Out goal is to help others to work out safer so they don't have to suffer the same fate.
What it does
Using machine learning, a camera feed is able to detect whether a workout is being preformed correctly. It tells the user whether they used good form during their set so that they know when to improve.
How we built it
With the help of Google's Teachable Machine, we were able to create a pose model that tracks the user's movements to see how they preformed. We then took the data from the model and displayed it on a website that was created with React. The website allows for a user friendly experience as they try to better their workout experience.
Challenges we ran into
A problem we had was connecting the model that was created by Teachable Machine to run with React. We also struggled getting React to be hosted on GitHub pages.
Accomplishments that we're proud of
We created a react website that is able to host machine learning models that can help people stay safe.
What we learned
We learned how to use Teachable Machine and Media Pipe to create machine learning models.
What's next for Spot Something IDK
Next steps are train a better model and to add additional workouts to the website.
Built With
- javascript
- kintone
- python
- react
Log in or sign up for Devpost to join the conversation.