Real Time Posture Detector using MediaPipe and OpenCV
Sukhdeep Singh, Trilok Singh, Tarun Singh, Diana Nagpal · 2024
This work presents the development of a smart gym system utilizing the pre-trained posture estimation model, MediaPipe and torsion angle. While exercising offers numerous health benefits, improper form can pose significant risks. Existing resources like trainers and instructional videos are not always readily accessible, leading individuals to adopt self-learning methods. However, self-learning carries the risk of adopting incorrect postures, potentially causing both short-term and long-term health problems. This system addresses this challenge by analyzing and tracking user movements and postures in real-time using OpenCV. The proposed system then provides immediate feedback on any identified postures through a display screen, allowing users to rectify their mistakes and ensure safe and effective exercise routines with user-friendly interface. This novel approach empowers individuals to prioritize safe exercise techniques and achieve the fitness goals.