Applying Sum and Max Product Algorithms of Belief Propagation to 3D Shape Matching and Registration

Author

Xiao, Pengdong ; Barnes, Nick ; Lieby, Paulette ; Caetano, Tiberio

Author_Institution

Res. Sch. of Inf. Sci. & Eng., Australian Nat. Univ., Canberra, ACT, Australia

fYear

2009

fDate

1-3 Dec. 2009

Firstpage

387

Lastpage

394

Abstract

3D shape matching based on meshed surfaces can be formulated as an energy function minimisation problem under a Markov random field (MRF) framework. However, to solve such a global optimisation problem is NP-hard. So research mainly focuses on approximation algorithms. One of the best known is belief propagation (BP), which has shown success in early vision and many other practical applications. In this paper, we investigate the application of both sum and max product algorithms of belief propagation to 3D shape matching. We also apply the 3D shape matching results to a 3D registration problem.

Keywords

Markov processes; approximation theory; belief networks; image matching; image registration; minimisation; 3D registration; 3D shape matching; Markov random field; NP-hard problem; approximation algorithms; belief propagation; energy function minimisation problem; max product algorithm; optimisation; sum product algorithm; Australia; Belief propagation; Computer applications; Databases; Digital images; Markov random fields; Minimization methods; Power engineering and energy; Shape; State-space methods; belief propagation; max product; shape matching; sum product;

fLanguage

English

Publisher

ieee

Conference_Titel

Digital Image Computing: Techniques and Applications, 2009. DICTA '09.

Conference_Location

Melbourne, VIC

Print_ISBN

978-1-4244-5297-2

Electronic_ISBN

978-0-7695-3866-2

Type

conf

DOI

10.1109/DICTA.2009.70

Filename

5384933