Classification of Human Poses Using Deep Learning Techniques

In numerous applications, such as tracking physical activity, decoding sign language, and controlling full-body gestures, human position estimation from the video is crucial. Human pose classification is challenging yet well-researched area. The particular way that human joints are arranged is referred to as pose. The localization of human joints or predetermined markers in images and videos can therefore be defined as the topic of “Human Pose Estimation.” Pose estimate is required for applications like human activity detection, sports, yoga positions, sign language applications, motion. The Mediapipe API is utilized in this study for human classification. This model categorizes human poses while identifying and tracking the actions of individual body parts. This paper demonstrates the adaptability of deep learning techniques and highlights the efficiency and accuracy of the BlazePose API in providing reliable pose landmarks.