Towards a local Kalman filter for visual tracking

This paper considers the task of closed curve filtering for visual tracking. Segmentation-based visual tracking strategies provide the closed curve measurements to filter. This paper discusses the derivation of a local, linear description for planar curve variation and curve uncertainty. It consists of a family of non-intersecting trajectories transverse to a given curve. Along one of the single-dimensional transverse trajectories, linear curve operations are feasible. Using the linear operation, a simple, locally optimal filtering procedure is derived. The filtering procedure is then used to define an observer for segmentation-based visual tracking. Experiments conducted validate the proposed method and resulting observer design.