Yoga Pose Detection Using Deep Learning

DOI

No DOI

Published In

Volume 3, Issue 6 (2022)

Published Online

November-December 2022

Pages

92-94

Article Type

Research Article

Check for updates

No DOI available for cross-reference

Abstract: India has a long history of being connected to yoga, an age old art form .It aids inpromoting physical fitness and at the same time offers mental tranquilly. Yoga is challenging topractice now that Covid-19 has been released. Class and might result in catastrophic injury ifundertaken without instruction. We create a system, here that detected the many yoga poses thatuser have adopted .The system make use of open-source information with six different videos of yoga poses done by 15 different volunteers. First, the goes through two processes to extract the data. The second phase the use of points data from the video dataset using the media pipe pose estimation library. Employing classification-based machine learning to pre-process the acquired data, train the data and test the data algorithms. Logistic regression, support vector machine classifier random forest classifier, k closest neighbors classifier, and naïve Bayes classifier the some of the machine learning algorithms employed. The method scores a 94% accuracy rate and the system is designed to operate on still photos, static movies, and live videos with a threshold value, below which the solution is rejected.