We all know the importance of adopting a healthy lifestyle. Besides keeping your body energized, fitness can also help achieve mental calmness. During the pandemic staying fit was one of the major concerns of many individuals and because of the lockdown and quarantines our whole lifestyle got affected. Another major concern was the discomfort and injuries caused when a workout routine is done in a wrong way. WorkoutRight gives you the tools, the motivation, and the support you need to stay fit.
What it does
WorkoutRight is a ML assisted personal trainer. The main idea was to monitor a person while he/she is working out and give them feedback on their performance. Broadly how well he/she is performing that particular exercise. So we thought of dividing the project into two sections, one for the rep-based exercises and the other for Yoga.
In the workout section, the number of reps that you perform of a particular exercise is counted and in the Yoga section, you are assisted in performing a Yoga exercise correctly with the help of markers tracked over your body which turn green when that particular body part is in right posture and at the correct angle for that Yoga Pose. Another thing to note is that with this approach of ML-assisted fitness, several discomforts and injuries which are caused due to an exercise performed in the wrong posture can be minimized.
How we built it
Built with Google's open-source machine learning solution MediaPipe and ReactJS library.
Challenges we ran into
- One of the major challenge that we faced was to track a person's body in real time and make sure the assessment was precise and fast enough.
- Another challenge was to make the project optimized enough so that the overall user experience is smooth and seamless.
- Our team didn't know much about web development so we had to learn a couple of new things to make it a web based project.
Accomplishments that we're proud of
- We were able to achieve all of the major functionalities that we planned beforehand.
- Our project motivates people to stay fit without dealing with the hassle of going to the gym or hiring a personal trainer.
What we learned
Working remotely and virtually as a team is one big thing that we all learnt during this Hackathon. On the technical aspect we learnt many concepts of Web Development, Machine learning and Computer Vision.
What's next for Workout Right
- Integration of physical devices like smart watches and fit bands into the project so that we can even monitor a persons BP, HeartRate and Oxygen level through out the workout session.
- Making the project production ready and more scalable.