<![CDATA[Post-pandemic lifestyle patterns have undergone many changes with the implementation of digital transformation, one of which is the meditation pattern such as yoga practice that can be done independently at home without direct interaction with the instructor. This study also aims to develop a yoga movement classification system using Convolutional Neural Network (CNN) based on human posture correction. Using the Movenet model, this system can recognise and classify different yoga poses to provide accurate feedback on correct posture. Training data was collected from yoga photographs and processed into pose images that were analysed using CNN. The results of this study indicate that the developed system is able to achieve a high level of accuracy in identifying yoga poses, which has the potential to help users improve their posture and reduce the risk of injury. This system is also implemented in a mobile application, making it easier for users to access posture correction in real time. As such, this research makes a significant contribution to the fields of health and technology by providing innovative solutions for safer and more effective yoga practice.]]>
Copyrights © 2024
