Wormholes in shape space: tracking through discontinuous changes in shape
- 27 November 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 344-349
- https://doi.org/10.1109/iccv.1998.710741
Abstract
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.Keywords
This publication has 7 references indexed in Scilit:
- Towards 3D hand tracking using a deformable modelPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- An efficient method for contour tracking using active shape modelsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- A framework for spatiotemporal control in the tracking of visual contoursInternational Journal of Computer Vision, 1993
- A Generic System for Image Interpretation Using Flexible TemplatesPublished by British Machine Vision Association and Society for Pattern Recognition ,1992
- Training Models of Shape from Sets of ExamplesPublished by Springer Nature ,1992
- Active Shape Models — ‘Smart Snakes’Published by Springer Nature ,1992
- Recovery of nonrigid motion and structurePublished by Institute of Electrical and Electronics Engineers (IEEE) ,1991