Yoga Pose Classification using Angle Heuristic Approach

Pose estimation plays vital role in various applications related to fitness. Yoga aims to development of all-round personality. In the past, it is used to be performed under the supervision of expert. In this fast and continuously changing world, yoga asanas can be performed utilizing the presented solution in this research work. Here, yoga asanas are classified by using Mediapipe and Angle heuristic approach. Mediapipe employs Blazepose model to identify 33 body keypoints, which are more than the number of points determined by conventional approaches of keypoint detection methods.