This study aimed to employ computer vision and deep learning methods in order to capture skeleton push start kinematics. Push start data were captured concurrently by a marker-based motion capture system and a custom markerless system. Very good levels of agreement were found between systems, particularly for spatial based variables (step length error 0.001 ± 0.012 m) while errors for temporal variables (ground contact time and flight time) were within 1.5 frames of the criterion measures. The computer vision based methods tested in this research provide a viable alternative to marker-based motion capture systems. Furthermore they can be deployed into challenging, real world environments to non-invasively capture data where traditional approaches would fail.
© Copyright 2020 ISBS Proceedings Archive (Michigan). Northern Michigan University. Published by International Society of Biomechanics in Sports. All rights reserved.