Temporal Smoothing for 3D Human Pose Estimation and Localization for Occluded People

In multi-person pose estimation actors can be heavily occluded, even becomefully invisible behind another person. While temporal methods can still predicta reasonable estimation for a temporarily disappeared pose using past andfuture frames, they exhibit large errors nevertheless. We present an energyminimization approach to generate smooth, valid trajectories in time, bridginggaps in visibility. We show that it is better than other interpolation basedapproaches and achieves state of the art results. In addition, we present thesynthetic MuCo-Temp dataset, a temporal extension of the MuCo-3DHP dataset. Ourcode is made publicly available.