A Yoga Pose Evaluation System for Enhanced Yoga Education

The significant increase in the cases of mental health illness in students instigated the integration of yoga in schools and universities due to its proven physical and mental health benefits. However, several challenges arise in these yoga classes because it is difficult for the instructors to objectively assess if their students are religiously and accurately performing the yoga poses being taught. The study aims to enhance yoga education in the academic setting by utilizing human pose estimation (HPE) to provide a way of objectively evaluating students' pose execution. An Android application was developed to perform real-time pose evaluation on two performance metrics, pose accuracy and pose consistency. The application extracts essential joint angles of the students during their performance and uses these angles to compute for accuracy and consistency. An inter-rater reliability test was performed to determine the degree of agreement between the system’s and instructor’s evaluations. The results generated kappa values of 0.6296 and 0.6429 for accuracy and consistency grading, respectively, indicating a substantial agreement between both evaluations. With this, the system can be used by instructors as a tool to assess their students’ pose accuracy and consistency objectively. Moreover, the records of the evaluations of the system can serve as a basis for the students’ final grades. Finally, the system has potential applications in research and training, enabling standardized and consistent evaluations of yoga practitioners across different settings.