With the rise of remote work and home workouts, there is an increasing need for a cost-effective, easy-to-use solution for tracking exercise form and count. Traditional methods require manual counting, which is prone to errors and distractions.
The project aims to create an automated, real-time exercise detection and counting system that utilizes computer vision to monitor and count five types of exercises: Push-ups, Bicep Curls, Squats, Sit-ups, and Lunges. The software provides immediate visual feedback, allowing users to improve their form and track their progress.
• Python • OpenCV (Open Source Computer Vision Library) • MediaPipe by Google • NumPy
• The program can successfully detect and count the five different exercises with a high degree of accuracy. • Needs adequate lighting and simple background for best results. • The system can be further enhanced by adding more exercises and integrating machine learning algorithms for more complex pose detection.
https://github.com/heathbrew/Real-time-Exercise-Detection-and-Counter-for-Home-Workouts