Real Time Multiple People Tracking and Pose Estimation

摘要

In this paper we present a combined probability estimation approach to detect and track multiple people for pose estimation at the same time. It can deal with partial and total occlusion between persons by adding torso appearance to the tracker. Moreover, the upper body of each individual is further segmented into head, torso, upper arm and lower arm in a hierarchical way. The simplicity of the feature and the simplified model allow close real time performance of the tracker. The experimental results show that the proposed method can deal with most of the inner-occlusion between persons, as well as certain self-occlusion. It's also much faster than the existing methods with comparable accuracy.