Skip to Main Content
This paper addresses the problem of determining a person's body and head orientations while tracking the person in an indoor environment monitored by a single top-view camera. The challenging part of this problem lies in the wide range of human postures depending on the position of the camera and articulations of the pose. In this work, a two-level cascaded particle filter approach is introduced to track humans. Color clues are used as the first level for each iteration and edge-orientation histograms are reutilized to support the tracking at the second level. To determine body and head orientations, a combination of Shape Context and SIFT features is proposed. Body orientation is calculated by matching the upper region of the body with predefined shape templates, then finding the orientation within the ranges of Â¿/8 degrees. Then, the optical flow vectors of SIFT features around the head region are calculated to evaluate the direction and type of the motion of the body and head. We demonstrate preliminary results of our approach showing that body and head orientations are successfully estimated. A discussion on various motion patterns and future improvements for more complicated situations is also given.