Yoga Pose Estimation Using YOLO Model

Abstract

Yoga, an historic exercise for every person and intellectual nicely-being, has gained large recognition in todays world . Monitoring yoga postures is essential for ensuring proper form, but yoga without an yoga professional is not good for person. In this research, we go deeply into the area of video processing ,image processing and deep learning, providing a new approach to yoga pose estimation. Leveraging the modern day YOLO (You Only Look Once) pose version, we suggest an revolutionary answer for real-time yoga pose detection in both images and video. The YOLO structure is customized and adaptative to recognizing and detecting yoga postures, maximizing accuracy and performance. YOLO is a popular deep learning algorithm used for object detection and classification in images and videos. It's known for its speed and efficiency. We inspect the result with in the parameter and also use customize data. We specially design this model for yoga Beginner and Intermediate those who cannot afford personal trainer or not able to mange time. Additionally, we renowned the limitations encountered in the course of our studies, also covering the way for future investigations. Keywords: Pose Estimation,YOLO,KeyPoints Detection,Virtual yoga ,ML models