A View Independent Classification Framework for Yoga Postures

TL;DR

This paper introduces a view-independent yoga posture classification framework using transfer learning from human pose estimation models, evaluated on diverse datasets and scenarios to ensure robustness and generalizability.

Contribution

It presents a novel three-step evaluation scheme for assessing yoga posture classifiers across unseen frames, subjects, and camera angles, emphasizing real-world applicability.

Findings

Transfer learning improves classification accuracy.

Validation accuracy varies significantly with cross-validation methods.

Classifier performance is robust across different camera angles and subjects.

Abstract

Yoga is a globally acclaimed and widely recommended practice for a healthy living. Maintaining correct posture while performing a Yogasana is of utmost importance. In this work, we employ transfer learning from Human Pose Estimation models for extracting 136 key-points spread all over the body to train a Random Forest classifier which is used for estimation of the Yogasanas. The results are evaluated on an in-house collected extensive yoga video database of 51 subjects recorded from 4 different camera angles. We propose a 3 step scheme for evaluating the generalizability of a Yoga classifier by testing it on 1) unseen frames, 2) unseen subjects, and 3) unseen camera angles. We argue that for most of the applications, validation accuracies on unseen subjects and unseen camera angles would be most important. We empirically analyze over three public datasets, the advantage of transfer learning and the possibilities of target leakage. We further demonstrate that the classification accuracies critically depend on the cross validation method employed and can often be misleading. To promote further research, we have made key-points dataset and code publicly available.