XRSZE

We can't spell exercise, but we can (probably) get you to exercise.

Comment

Inspiration 🌟

Exercising is one of the many elements to a balanced and healthy lifestyle, especially in our profession as some of us tend to slack in it. We want XRSZE to make you get off your butt and do things!

What it does 📝

XRSZE counts your reps for various exercises while keeping you accountable for your fitness goals. We also have an AI Chatbot (made using Groq) that can give you a specialized meal plan, workout routine and even motivation for your specific height, weight, age and dietary restrictions!

How we built it 🛠️

We built XRSZE using various web technologies and APIs. The front-end interface was designed with HTML/CSS/JS for simplistic beauty. For the backend, we used Node.JS and integrated it with MediaPipe for the Computer Vision and movement tracking while using the Groq API for the Chabot. We fine-tuned our Groq AI Chatbot to specifically work greatly with fitness and food consumption and to give advice to improve your health. We also specifically calibrated the MediaPipe recognition model to detect accurate body movements to count and recognize workouts and count/time the activities accurately.

Challenges we ran into 😓

One of the biggest challenges was ensuring the accuracy of movement tracking with MediaPipe. Fine-tuning the model to reliably count and recognize various exercises required extensive testing and adjustments. We also faced difficulties integrating the Groq API seamlessly with our backend, which led to several iterations before achieving smooth interaction between the chatbot and user inputs. Additionally, optimizing the front-end performance to handle real-time data without lag was crucial for a smooth user experience.

What we learned 📚

We learned a lot about integrating complex technologies, like MediaPipe and Groq, to work together in a unified system. The process taught us the importance of iterative testing and refinement, especially when fine-tuning AI models for specific applications. We also gained valuable insights into optimizing real-time data processing and enhancing user experience through thoughtful design and responsive feedback mechanisms. It was also some of our first times using Node.JS and it was a great learning experience to learn and improve in real time.

What's next for XRSZE 🚀

We’re excited to expand XRSZE into a mobile app, and we are already working on the Flutter code to make our platform more accessible and convenient for users on the go. We also plan to enhance our workout tracking capabilities by adding support for a wider variety of exercises and workouts, ensuring that users have a comprehensive tool to meet their fitness goals.

Share this project:

Updates