Body Joint Detection Analysis And Human Pose Classification Using Cnn A Case Study On Yoga Poses

Abstract

Human pose estimation (HPE) is one of the thrust area of computer vision in which pose is estimated by identifying the correct position of human body joints using machine learning and deep learning approaches. HPE is the process of estimating the location of the human body joints from the image or video. The correct estimate of human body joints has been used to track people s minimal activities in real-time applications. A novel approach to object detection focusing on identifying people by utilizing bounding boxes. However, pose detection and pose tracking method enable computers to understand human body language. HPE is an extensive research area that relies on multiple individuals being monitored. newlineVirtual motion and pose from images and video can be estimated by detecting body joints and their interconnection. The human body has diverse and complicated poses in yoga, making its classification challenging. This study estimates yoga poses from the images using a neural network. Five different yoga poses, viz. Downdog, Tree, Plank, Warrior2, and Goddess in the form of RGB images and used as the target inputs. The Blazepose model is used to localize the body joints of the yoga poses. It can detect a maximum of 33 body joints, referred to as keypoints, covering almost all the body parts. Keypoints achieved through the model have been considered as predicted joint locations. True keypoints as the ground truth body joint for individual yoga poses have been identified manually using the open source Image annotation tool named Makesense AI. newlineA detailed analysis of the body joint detection accuracy is proposed in the form of Percentage of Corrected Keypoints (PCK) and Percentage of Detected Joints (PDJ) for individual body parts and individual body joints, respectively. An algorithm is designed to measure PCK and PDJ in which the distance between the predicted joint location and true joint location can be calculated. The experiment evaluation suggests that the adopted model succeed to get 93.9% PCK for Goddess pose. The