Applying sum and max product algorithms of belief propagation to 3D shape matching and registration

Pengdong Xiao; Nick Barnes; Paulette Lieby; Tiberio Caetano

doi:10.1109/DICTA.2009.70

Applying sum and max product algorithms of belief propagation to 3D shape matching and registration

Pengdong Xiao^*, Nick Barnes, Paulette Lieby, Tiberio Caetano

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Citation (Scopus)

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.

Original language	English
Title of host publication	DICTA 2009 - Digital Image Computing
Subtitle of host publication	Techniques and Applications
Pages	387-394
Number of pages	8
DOIs	https://doi.org/10.1109/DICTA.2009.70
Publication status	Published - 2009
Event	Digital Image Computing: Techniques and Applications, DICTA 2009 - Melbourne, VIC, Australia Duration: 1 Dec 2009 → 3 Dec 2009

Publication series

Name	DICTA 2009 - Digital Image Computing: Techniques and Applications

Conference

Conference	Digital Image Computing: Techniques and Applications, DICTA 2009
Country/Territory	Australia
City	Melbourne, VIC
Period	1/12/09 → 3/12/09

Access to Document

10.1109/DICTA.2009.70

Cite this

@inproceedings{29ad547e94aa4ee489e1dc0569865e55,

title = "Applying sum and max product algorithms of belief propagation to 3D shape matching and registration",

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 = "Belief propagation, Max product, Shape matching, Sum product",

author = "Pengdong Xiao and Nick Barnes and Paulette Lieby and Tiberio Caetano",

year = "2009",

doi = "10.1109/DICTA.2009.70",

language = "English",

isbn = "9780769538662",

series = "DICTA 2009 - Digital Image Computing: Techniques and Applications",

pages = "387--394",

booktitle = "DICTA 2009 - Digital Image Computing",

note = "Digital Image Computing: Techniques and Applications, DICTA 2009 ; Conference date: 01-12-2009 Through 03-12-2009",

}

Xiao, P, Barnes, N, Lieby, P & Caetano, T 2009, Applying sum and max product algorithms of belief propagation to 3D shape matching and registration. in DICTA 2009 - Digital Image Computing: Techniques and Applications., 5384933, DICTA 2009 - Digital Image Computing: Techniques and Applications, pp. 387-394, Digital Image Computing: Techniques and Applications, DICTA 2009, Melbourne, VIC, Australia, 1/12/09. https://doi.org/10.1109/DICTA.2009.70

Applying sum and max product algorithms of belief propagation to 3D shape matching and registration. / Xiao, Pengdong; Barnes, Nick; Lieby, Paulette et al.
DICTA 2009 - Digital Image Computing: Techniques and Applications. 2009. p. 387-394 5384933 (DICTA 2009 - Digital Image Computing: Techniques and Applications).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Applying sum and max product algorithms of belief propagation to 3D shape matching and registration

AU - Xiao, Pengdong

AU - Barnes, Nick

AU - Lieby, Paulette

AU - Caetano, Tiberio

PY - 2009

Y1 - 2009

N2 - 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.

AB - 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.

KW - Belief propagation

KW - Max product

KW - Shape matching

KW - Sum product

UR - http://www.scopus.com/inward/record.url?scp=77950310439&partnerID=8YFLogxK

U2 - 10.1109/DICTA.2009.70

DO - 10.1109/DICTA.2009.70

M3 - Conference contribution

SN - 9780769538662

T3 - DICTA 2009 - Digital Image Computing: Techniques and Applications

SP - 387

EP - 394

BT - DICTA 2009 - Digital Image Computing

T2 - Digital Image Computing: Techniques and Applications, DICTA 2009

Y2 - 1 December 2009 through 3 December 2009

ER -

Applying sum and max product algorithms of belief propagation to 3D shape matching and registration

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this