DocumentCode
2711750
Title
Factorized graph matching
Author
Zhou, Feng ; De La Torre, Fernando
Author_Institution
Robot. Inst., Carnegie Mellon Univ., Pittsburgh, PA, USA
fYear
2012
fDate
16-21 June 2012
Firstpage
127
Lastpage
134
Abstract
Graph matching plays a central role in solving correspondence problems in computer vision. Graph matching problems that incorporate pair-wise constraints can be cast as a quadratic assignment problem (QAP). Unfortunately, QAP is NP-hard and many algorithms have been proposed to solve different relaxations. This paper presents factorized graph matching (FGM), a novel framework for interpreting and optimizing graph matching problems. In this work we show that the affinity matrix can be factorized as a Kronecker product of smaller matrices. There are three main benefits of using this factorization in graph matching: (1) There is no need to compute the costly (in space and time) pair-wise affinity matrix; (2) The factorization provides a taxonomy for graph matching and reveals the connection among several methods; (3) Using the factorization we derive a new approximation of the original problem that improves state-of-the-art algorithms in graph matching. Experimental results in synthetic and real databases illustrate the benefits of FGM. The code is available at http://humansensing.cs.cmu.edu/fgm.
Keywords
approximation theory; computational complexity; computer vision; image matching; matrix decomposition; quadratic programming; visual databases; Kronecker product; NP-hard problem; approximation; computer vision; factorized graph matching; graph matching problem interpretation; graph matching problem optimization; matrix factorization; pair-wise affinity matrix; pair-wise constraint; quadratic assignment problem; real database; synthetic database; Approximation algorithms; Computer vision; Linear approximation; Linear programming; Optimization; Vectors;
fLanguage
English
Publisher
ieee
Conference_Titel
Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on
Conference_Location
Providence, RI
ISSN
1063-6919
Print_ISBN
978-1-4673-1226-4
Electronic_ISBN
1063-6919
Type
conf
DOI
10.1109/CVPR.2012.6247667
Filename
6247667
Link To Document