Pose Estimation using Media Pipe

This project endeavors to develop a sophisticated Gym Curl Counter utilizing the capabilities of Python, MediaPipe, and machine learning techniques. The aim is to create a robust system that accurately tracks and counts the number of curls performed during gym workouts. The system begins by setting up MediaPipe, a powerful framework for building machine learning pipelines for various perception tasks. Using MediaPipe, the project estimates human poses in real-time video streams, enabling the extraction of joint coordinates from key points on the body. Subsequently, the project calculates the angles between specific joints, crucial for determining the curling motion during exercises. Leveraging Python’s mathematical libraries, the system precisely computes these angles, providing insights into the curling activity. Through the integration of machine learning algorithms, the Gym Curl Counter effectively distinguishes between curling and other movements, ensuring accurate counting of curls. By training the model on diverse datasets encompassing a range of curl variations, the system achieves high accuracy and reliability. In practical application, users can benefit from real-time feedback on their curling performance, including the number of repetitions completed and the quality of each curl. Additionally, the system can offer insights into form and technique, enhancing workout efficiency and reducing the risk of injury. Overall, by combining Python programming, MediaPipe, and machine learning methodologies, the Gym Curl Counter presents an innovative solution for monitoring and optimizing gym workouts, empowering individuals to achieve their fitness goals effectively and safely.