Healthy Living Facilitated by Deep Learning Algorithm
Rajeshri Pravin Shinkar · Advances in IT standards and standardization research (AISSR) book series/Advances in IT standards and standardization research series · 2024
Deep Learning is a subset of Machine Learning, Artificial Intelligence and Data Science. Deep Learning uses artificial neural networks for solving the complex problems. This paper gives a computer-vision based system for the real time posture of the exercise by using the classification of video streams. The system introduces Mediapipe, to perform the pose estimation and identify body landmarks for various types of exercise. This paper gives the performance of pose estimation and identify body landmarks with the help of Mediapipe and Convolutional Neural Network, by calculating key joint angles from the detected landmarks, the program classifies various yoga poses in real-time. The researcher describes the potential benefits for yoga practitioners by providing immediate feedback on their posture accuracy during practice, promoting self-correction and potentially leading to improved form and reduced injury risk.