Inspiration
We were inspired to create this project during online classes. We found that we couldn't do physical classes such as Acrobatics properly in an online setting. My coaches struggled to fix my posture online. Also after a while we found that we struggled to maintain correct posture during yoga. It was then when we had an "Eureka!" moment with the idea of this app.
What it does
The app contains an array of different courses ranging from yoga to even basketball. You can select a certain course and get started. It uses your camera and machine learning technology to compare your pose to that of the instructor; thus it can ensure that your movements and posture are correct.
How we built it
We used Google's Flutter kit as our base for this project and used dart as a primary language. We use the Google MLKit Technology for the imagine processing of the app. We used Firebase for our database and log in features and used Figma UI to think out the project blueprint itself.
Challenges we ran into
Flutter changed significantly since we last used it so it took a while to get used to the new documentation. For example: Instead of this:
document.data()['ID']
We now had to do this:
document.data()['ID']
(document.data()as dynamic)['Name']
It took us quite a while to get used to the new syntax.
Another problem we faced had to do with machine learning. Originally, we decided to use TensorFlow's Machine Learning kit in our program. However we soon found out that TensorFlow wasn't as compatible with dart and Flutter as we hoped. Hence, we decided to change to Google's ML Kit for the processing of our app.
Accomplishments that we're proud of and what we learned
It was extremely gratifying once we finally got the hang of the new flutter documentation. We believe that we now have a hang of it and use it with much more ease.
We are also very proud of the UI/UX of the app. We put a lot of effort into planning it with Figma UI and we are very proud of how it turned out.
What's next for FitAI
We are focussed on adding new courses for FitAI including some that our users may recommend.
We are also thinking of adding Wikis for each course to give the users an in-depth view of what they might sign up for
Built With
- dart
- figma
- firebase
- flutter
- machine-learning
- ml
- mlkit

Log in or sign up for Devpost to join the conversation.