Convolutional Neural Network Algorithm for Yoga Pose Estimation System as an Exercise Assistant

Yoga is a physical activity that includes meditation using stretching techniques, breathing, balance, and physical flexibility to achieve harmony and harmony between the emotions, soul, mentality, spirituality, and body. Yoga is a popular sport that can make your body and mind healthy and strong. Artificial intelligence technology can be used to gain new experiences in yoga. A healthy lifestyle for people who have changed after the pandemic. Therefore, this study aims to utilize artificial intelligence technology to support public awareness of maintaining a healthy lifestyle by practicing yoga independently. This study uses the Convolutional Neural Network (CNN) algorithm to classify yoga pose images as exercise assistants that can show whether the yoga poses are appropriate. With a total of 2672 image data containing seven types of yoga pose images, the results show that the CNN algorithm can detect and classify yoga poses well by obtaining an accuracy of 98.91%.