Assessing Performance of Aerobic Routines using Background Subtraction and Intersected Image Region

Abstract

It is recommended for a novice person to engage trained personnel before starting an unfamiliar aerobic or weight routine to gain real-time expert feedbacks. This greatly reduces the risk of injury and maximise physical gains. We present a simple image similarity measure based on intersected image region to assess a subject's performance of an aerobic routine. The method was implemented inside an Augmented Reality (AR) desktop app that employed a single RGB camera to capture still images of the subject as he or she progressed through the routine. The background-subtracted body pose image was compared against the exemplar image (i.e., AR template) at specific intervals. Based on a limited dataset, our pose matching function managed an accuracy of 93.67%.