The purpose of this study is to develop a method for quantitative evaluation and visualization of inter-trial differences in the motion of athletes. Previous methods for kinematic analyses of human movement have required attaching specific equipment to a body segment or can only be used in an environment designed for analyses. Therefore, they are difficult to use for observing motions in real games. To enhance the applicability to real-game situations, we propose appearance-based motion tracking. Our method only requires an image sequence from a camera. From the image sequence, automatic detection of trials and a difference analysis of them are conducted. We applied our method to the analysis of pitching motions in actual baseball games. Though we have no quantitative evaluations yet, the experimental results imply the efficacy of our method.
© Copyright 2012 ISBS - Conference Proceedings Archive (Konstanz). Springer. Veröffentlicht von International Society of Biomechanics in Sports. Alle Rechte vorbehalten.