Core

Float out of your comfort zone.

Comment

Inspiration

Feeling major self-doubt when you first start hitting the gym or injuring yourself accidentally while working out are not uncommon experiences for most people. This inspired us to create Core, a platform to empower our users to take control of their well-being by removing the financial barriers around fitness.

What it does

Core analyses the movements performed by the user and provides live auditory feedback on their form, allowing them to stay fully present and engaged during their workout. Our users can also take advantage of the visual indications on the screen where they can view a graph of the keypoint which can be used to reduce the risk of potential injury.

How we built it

Prior to development, a prototype was created on Figma which was used as a reference point when the app was developed in ReactJs. In order to recognize the joints of the user and perform analysis, Tensorflow's MoveNet model was integrated into Core.

Challenges we ran into

Initially, it was planned that Core would serve as a mobile application built using React Native, but as we developed a better understanding of the structure, we saw more potential in a cross-platform website. Our team was relatively inexperienced with the technologies that were used, which meant learning had to be done in parallel with the development.

Accomplishments that we're proud of

This hackathon allowed us to develop code in ReactJs, and we hope that our learnings can be applied to our future endeavours. Most of us were also new to hackathons, and it was really rewarding to see how much we accomplished throughout the weekend.

What we learned

We gained a better understanding of the technologies used and learned how to develop for the fast-paced nature of hackathons.

What's next for Core

Currently, Core uses TensorFlow to track several key points and analyzes the information with mathematical models to determine the statistical probability of the correctness of the user's form. However, there's scope for improvement by implementing a machine learning model that is trained on Big Data to yield higher performance and accuracy. We'd also love to expand our collection of exercises to include a wider variety of possible workouts.

Built With

Share this project:

Updates