Real-Time Estimation of Missing Markers for Reconstruction of Human Motion

Abstract

Optical motion capture is a prevalent technique for capturing and analyzing movement. However, a common problem in optical motion capture is the missing marker problem due to occlusions or ambiguities. Most methods for resolving this problem either require extensive post-processing efforts or become ineffective when a significant portion of markers are missing for extended periods of time. In this paper, we present an approach to reconstruct human motion corrupted in the presence of missing or mis-tracking markers. We propose a data-driven, piecewise linear predicting kalman filter framework to estimate missing marker position, and reconstruct human motion in real time by rigid body tracking solver. It allows us to accurately and effectively reconstruct human motion within a simple extrapolation framework. We demonstrate the effectiveness of our method on real motion data captured using OptiTrack. Our experimental results demonstrate that our method is efficient in recovering human motion.