VitalityAI

First time Hackathon, Scenario 2, Best Front End

Inspiration

While deciding on an idea, we thought "what is the most important thing throughout every major life change?" The answer was simple: your health. We often get so caught up in life's demands that we neglect our health. VitalityAI aims to solve this problem, it acts as an accountability partner to keep your health in check, no matter how busy we get.

What it does

VitalityAI is a mobile fitness coach that acts as a workout partner. It counts your reps for you and keeps track of your progress. You tell it your goals, and it keeps track of them for you. It gamifies fitness so that it feels like less of something you have to do, but something that's accomplishing to do.

How we built it

VitalityAI was built using React Native with Expo and Javascript. We used TensorFlow Lite's Movenet for real-time pose estimation.

Challenges we ran into

Using the pose estimation model proved challenging when trying to calibrate movements accurately, and the rep counter was a nightmare to perfect. Styling in react-native with so many components took a ton of time from providing actual functionality. We used some aspects of react-native that we had to learn during the hackathon, such as using modals.

Accomplishments that we're proud of

We spent a lot of time perfecting the UI and theme, which we think turned out exactly as we envisioned it. Successfully integrating the camera with the pose-estimation model and it tracking each movement accurately was also a huge breakthrough for us. We also believe that setting goals is one of the most important aspects of improving any part of your life, and implementing the goal tracker was a vision come true.

What we learned

We learned how to split up a workload as a team effectively, in an approach where we start with our own tasks and then integrate. Time management was a huge problem for us, but we learned how to set time estimations for our workloads effectively. We had a lot of ideas we wanted to implement, but we learned to prioritize the core features.

What's next for VitalityAI

We are planning to add more features such as a photo gallery and a broader exercise selection. We also want to improve the accuracy of the model so that we can release it to the app store. VitalityAI is made for everyone and can be used by everyone to keep their fitness in check.

Built With

• javascript
• movenet
• react-native
• tensorflow-lite