Doing Squat

• 

  Dashboard

• 

  Activity Log Summary

• 

  Activity Log

• 

  Exercise Selection

• 

  Exercise Groups

Inspiration

We wanted to create a solution for helping students and professionals stay accountable for their productivity time while encouraging them to engage in a healthy lifestyle.

What it does

Users are able to choose apps that they can lock access to, only allowing time-limited access with a bank of "reward" minutes that are earned from completing fitness challenges. Our app tracks fitness challenges by relying on device movement, providing a convenient way for users to measure activity without the inconvenience of traditional camera based computer vision setups.

How we built it

We wrote the application using React and deployed to mobile using Expo. Python was while training the machine learning model we utilized to measure user movement.

Challenges we ran into

By relying only on device movement to extrapolate bodily movement, we had to find a way to measure body movement using only device movement. Using a machine learning model to extrapolate body movement from device movement proved to be a challenge due to the constraints of the device measurements.

Accomplishments that we're proud of

This was our first time developing and deploying a mobile application as a team. We learned a lot about using a React Native framework (Expo) to program a cross-platform mobile app, and the processes in training and integrating machine learning models.

What we learned

Not to attempt to train a machine learning model (for the first time) in less than 24 hours.

What's next for Doing Squat

We'd like to complete a full implementation of the app-locking feature using iOS's Screen Time features and using the relative Android implementations. This way, reward minutes could be utilized in actual app access for users. We'd also like to offer additional exercises that are able to be movement tracked.

Built With

• expo.io
• python
• react
• typescript