Wormholes in shape space: tracking through discontinuous changes in shape

Existing object tracking algorithms generally use some form of local optimisation, assuming that an object´s position and shape change smoothly over time. In some situations this assumption is not valid: the track able shape of an object may change discontinuously, for example if it is the 2D silhouette of a 3D object. In this paper we propose a novel method for modelling temporal shape discontinuities explicitly. Allowable shapes are represented as a union of (learned) bounded regions within a shape space. Discontinuous shape changes are described in terms of transitions between these regions. Transition probabilities are learned from training sequences and stored in a Markov model. In this way we can create `wormholes´ in shape space. Tracking with such models is via an adaptation, of the CONDENSATION algorithm