SquatSense - Computer Vision Personal Trainer

Bad form being detected along with Amazon echo giving feedback
Good form being detected along with Amazon echo giving motivation!

We’ve all been there—standing in the gym, unsure if our squat form is correct, but too intimidated to ask for help. The fear of injury, the struggle to get real-time feedback, along with the hesitation to approach a trainer led us to one question:

What if you had an computer vision powered personal trainer, right in your home, guiding you every step of the way?

That’s how SquatSense was born, an computer vision based fitness assistant that ensures every squat counts, every movement is safe, and every workout is effective.

--> What it does SquatSense combines the power of Computer Vision, real-time feedback correction, and smart home integration with Alexa echo to boost your confidence in workouts:

✔ Computer Vision– Tracks your hip, knee and ankle alignment, and posture using just a webcam. ✔ Instant Visual Feedback – LED indicators provide real-time corrections (Green = Good, Red = Bad). ✔ Alexa Voice Coaching – Your AI trainer speaks to encourage and gives you specific techniques to correct your form while you train. ✔ Live Rep Counting tracking– No more guessing—stay focused and track your progress seamlessly.

No wearables. No expensive gym memberships. Just CV-powered personal training at your fingertips!

--> How we built it We combined computer vision, hardware integration, and smart home automation into one seamless system:

✔ AI-Powered Motion Tracking: Using OpenCV & Mediapipe, we detect squat form, track key joints, and analyze movement in real-time. ✔ Smart Feedback System: An Arduino Uno controls LEDs and a display, signaling squat quality instantly. ✔ Real-Time Communication: PySerial bridges Python and Arduino, ensuring accurate data transmission. ✔ Alexa Integration via IFTTT: Our AI trainer talks to you—offering real-time coaching and corrections through Amazon Echo.

What makes SquatSense special? there's no inconvenience of extra sensors, no wearables—just your webcam, WiFi, and AI-powered coaching.

--> Challenges we ran into Building SquatSense wasn’t easy at all. We faced:

Challenges getting Alexa to deliver instant squat feedback - it was a huge hurdle.
Hardware Communication and ensuring Arduino + Python worked in sync for real-time LED updates.

--> Accomplishments that we're proud of ✔ We built a fully functional AI personal trainer that works with just a webcam! ✔ Seamless integration between Computer Vision, Arduino, and Alexa. ✔ Overcame major technical challenges in smart home automation and real-time feedback. ✔ Created an accessible and intuitive solution for fitness enthusiasts at any level.

--> What we learned 🔹 AI + Fitness = Future of Training – AI-driven form correction can revolutionize home workouts. 🔹 Hardware & Software Integration – Merging computer vision, Arduino, and smart assistants is challenging but rewarding. 🔹 User Experience Matters – Real-time feedback via lights, voice, and displays makes workouts engaging and effective.

--> What's next for SquatSense - AI Personal Trainer We’re just getting started! Here’s what’s next:

🔹Expanding to Full-Body Workouts – Push-ups, lunges, deadlifts, and more! 🔹Gamification & Challenges – Compete with friends, track progress, and earn fitness streaks. 🔹 Multi-Platform Support – Integration with mobile apps & smart mirrors. 🔹 AI Personalization – Adaptive coaching based on individual squat patterns.