RepTrack

https://www.canva.com/design/DAG5fIT7TXQ/HW0x99WSpTdW3hgKyaliwg/edit

Inspiration

We've all been there - trying to count reps during an intense workout, losing track mid-set, and having to guess whether we did 10 or 12 reps. Manual counting breaks your focus and leads to inaccurate tracking, which affects workout quality and progress measurement. We wanted to solve this universal fitness problem with an affordable, automatic solution that anyone could use.

What it does

RepTrack is a real-time workout rep counter that uses motion sensors (accelerometer and gyroscope) to automatically detect and count exercise repetitions. The system streams data to a live web dashboard where users can:

Track reps in real-time for both sensors
Monitor workout duration with a session timer
View intensity patterns for each rep through visual charts
Watch their form via live camera feed
Review past activity logs of sets and milestones

No more manual counting, no expensive wearables - just accurate, automatic rep tracking.

How we built it

Hardware: We used accelerometer and gyroscope motion sensors to capture exercise movements with high precision.

Backend: Built with C++ for real-time processing and low-latency rep counting. We used vcpkg for dependency management to handle the complex sensor data processing pipeline.

Transport Layer: Implemented WebSocket protocol for real-time bidirectional communication between the hardware sensors and the web interface.

Frontend: Created a responsive web dashboard using HTML, CSS, and JavaScript that displays live data with interactive charts and video feed.

Server: Node.js runtime handles WebSocket connections and manages data streaming between components.

Camera Integration: Live video feed for form monitoring and exercise validation. Using Presage to provide additional metrics

Challenges we ran into

Sensor Calibration: Getting the accelerometer and gyroscope to accurately detect different types of movements was difficult. We had to fine-tune sensitivity thresholds to avoid counting partial reps while still catching legitimate ones.

Real-time Data Streaming: Implementing low-latency WebSocket communication while maintaining data accuracy was challenging. We had to optimize our C++ processing pipeline to handle high-frequency sensor updates without lag.

Synchronization: Keeping the rep counts, timer, charts, and camera feed all synchronized in real-time required careful state management.

Cross-Platform Compatibility: Ensuring the web dashboard worked smoothly across different browsers and devices took significant testing and debugging.

Accomplishments that we're proud of

Built a fully functional real-time system in 36 hours that actually works and counts reps accurately
Successfully integrated hardware sensors with web technology using WebSocket protocol
Created a clean, intuitive dashboard that displays multiple data streams simultaneously
Implemented live camera feed alongside sensor data for comprehensive workout tracking
Achieved low latency processing - users see their rep counts update instantly

What we learned

How to work with IMU (Inertial Measurement Unit) sensors and process accelerometer/gyroscope data
Real-time data streaming with WebSocket protocol for bidirectional communication
C++ optimization techniques for high-frequency data processing
Signal processing to filter noise and detect patterns in motion data
How to integrate multiple data sources (sensors, camera, timer) into a cohesive dashboard
The importance of user feedback loops - seeing the rep count update in real-time makes the system intuitive

What's next for RepTrack

Multi-Exercise Recognition: Use machine learning to automatically detect which exercise you're doing (squats, curls, pushups, etc.) and apply exercise-specific counting algorithms.

Workout History & Analytics: Use our cloud-sync workout data to display progress over time with weekly/monthly trend analysis and personal records. Integration with popular fitness apps like MyFitnessPal and Apple Health.

AI-Powered Form Analysis: Combine camera feed with computer vision to provide real-time feedback on exercise technique - "keep your back straight," "go deeper on squats," etc. This would help prevent injuries and improve performance.

Social Features & Challenges: Add leaderboards, friend challenges, achievement badges, and streak tracking to gamify workouts and keep users motivated. Virtual workout sessions with friends for accountability.

Mobile App: Native iOS/Android app with Bluetooth connectivity to sensors for a seamless mobile experience.

Wearable Integration: Connect with Apple Watch, Fitbit, and Garmin devices to correlate heart rate data with rep counts for comprehensive fitness tracking.