Motion trajectories

Abstract

A simple algorithm for selecting and linking interesting flow vectors across a sequence of frames for computing motion trajectories is presented. Tokens that have both interesting pixel gray values in the spatial domain and in the optical flow field in the temporal domain are tracked. This AND operation effectively removes some redundant trajectories. Due to errors introduced during the computation of optical flow, and the linking of such flow vectors across a sequence of frames, the resultant trajectories are not always smooth. A Kalman-filtering-based approach for smoothing the trajectories is discussed.<>