• Uma Ade

he integration of Machine Learning (ML) and Computer Vision has opened new possibilities in fitness and health monitoring, enabling real-time analysis of human movements. This work presents a web-based fitness application that leverages ML-powered pose estimation techniques to assist users in performing exercises with proper form. The application uses live webcam inputs to detect key-points on the human body, analyse exercise poses, and provide instant feedback on the correctness of the user's posture. By tracking the user's exercise progress, the system also counts repetitions and evaluates performance. The proposed system employs advanced Computer Vision tools, such as OpenCV and MediaPipe’s BlazePose, for accurate pose detection and assessment. To address the health risks associated with improper exercise forms, such as injuries and muscle strain. Using a dataset comprising over 1,000 key-point coordinates for both correct and incorrect exercise postures, the system evaluates user movements and offers actionable corrective suggestions. This personalized feedback helps users improve their form and achieve better results. Designed to support a wide range of common exercises, the web application operates seamlessly on any device with a webcam, offering a cost-effective and accessible alternative to human trainers. By providing real-time feedback and progress tracking, the system empowers users to exercise safely and effectively in their own space. This work highlights the challenges and advancements in ML-based human pose estimation while showcasing the potential of AI-powered fitness tools to revolutionize personal health and fitness.

Computer Vision, Machine Learning, BlazePose,Pose detection, BlazePose, Health, AI-powered fitness,Workout,Human Pose Estimation.

"FITNESS WEB APPLICATION USING ML AND COMPUTER VISION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 1, page no.g321-g327, January-2025, Available :http://www.jetir.org/papers/JETIR2501633.pdf

"FITNESS WEB APPLICATION USING ML AND COMPUTER VISION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 1, page no. ppg321-g327, January-2025, Available at : http://www.jetir.org/papers/JETIR2501633.pdf