Motion characterization by temporal slices analysis

Abstract

This paper describes an approach to characterize camera and object motions based on the analysis of spatio-temporal image volumes. In the spatio-temporal slices of image volumes, motion is depicted as oriented patterns. We propose a tensor histogram computation algorithm to represent these patterns. The motion trajectories in a histogram are tracked to describe both the camera and object motions. In addition, we exploit the similarity of the temporal slices in a volume to reliably partition a volume into motion tractable units.