A Particle Filter Approach to Learning Partial Shape Correspondences

R. Lakaemper and M. Sobel (USA)


shape correspondence, object recognition, particle filters, partial matching


This paper describes a Particle Filter approach to solve the partial shape correspondence problem. The shapes are represented by ordered sequences of local features along their polygonal boundary. Learning domain knowledge which specifies additional global constraints, the Particle Filter system finds locally and globally consistent corre spondences between similar shape parts. Experiments us ing standard alignment techniques based on the given cor respondence relationships, demonstrate the advantages of this approach, outperforming related approaches in partial shape matching.

Important Links:

Go Back