Human Movement Science in The Wild: Can Current Deep-Learning Based Pose Estimation Free Us from The Lab?
L. Needham, M. Evans, D. P. Cosker, L. Wade, P. M. McGuigan, J. L. Bilzon, S. L. Colyer
This study evaluates deep-learning pose estimation accuracy against marker-based motion capture in real-world settings, finding systematic errors in joint center locations.
bioRxiv · Apr 22, 2021 · PreprintBioengineering