Outlier robust point correspondence based on GNCCP

Xu Yang; Hong Qiao; Zhi Yong Liu

doi:10.1016/j.patrec.2014.12.011

Outlier robust point correspondence based on GNCCP

Xu Yang
, Hong Qiao
, Zhi Yong Liu

Chinese Academy of Sciences

Research output: Contribution to journal › Article › peer-review

18 Scopus citations

Abstract

Graph matching is a fundamental problem in pattern recognition and computer vision. In this paper we introduce a novel graph matching algorithm to find the specified number of best vertex assignments between two labeled weighted graphs. The problem is first explicitly formulated as the minimization of a quadratic objective function and then solved by an optimization algorithm based on the recently proposed graduated nonconvexity and concavity procedure (GNCCP). Simulations on both synthetic data and real world images witness the effectiveness of the proposed method.

Original language	English
Pages (from-to)	8-14
Number of pages	7
Journal	Pattern Recognition Letters
Volume	55
DOIs	https://doi.org/10.1016/j.patrec.2014.12.011
State	Published - 1 Apr 2015
Externally published	Yes

Keywords

Feature correspondence
GNCCP
Graph matching

Access to Document

10.1016/j.patrec.2014.12.011

Cite this

@article{2d2c93cede354c2da7778b22a0eefbbe,

title = "Outlier robust point correspondence based on GNCCP",

abstract = "Graph matching is a fundamental problem in pattern recognition and computer vision. In this paper we introduce a novel graph matching algorithm to find the specified number of best vertex assignments between two labeled weighted graphs. The problem is first explicitly formulated as the minimization of a quadratic objective function and then solved by an optimization algorithm based on the recently proposed graduated nonconvexity and concavity procedure (GNCCP). Simulations on both synthetic data and real world images witness the effectiveness of the proposed method.",

keywords = "Feature correspondence, GNCCP, Graph matching",

author = "Xu Yang and Hong Qiao and Liu, \{Zhi Yong\}",

year = "2015",

month = apr,

day = "1",

doi = "10.1016/j.patrec.2014.12.011",

language = "英语",

volume = "55",

pages = "8--14",

journal = "Pattern Recognition Letters",

issn = "0167-8655",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Outlier robust point correspondence based on GNCCP

AU - Yang, Xu

AU - Qiao, Hong

AU - Liu, Zhi Yong

PY - 2015/4/1

Y1 - 2015/4/1

N2 - Graph matching is a fundamental problem in pattern recognition and computer vision. In this paper we introduce a novel graph matching algorithm to find the specified number of best vertex assignments between two labeled weighted graphs. The problem is first explicitly formulated as the minimization of a quadratic objective function and then solved by an optimization algorithm based on the recently proposed graduated nonconvexity and concavity procedure (GNCCP). Simulations on both synthetic data and real world images witness the effectiveness of the proposed method.

AB - Graph matching is a fundamental problem in pattern recognition and computer vision. In this paper we introduce a novel graph matching algorithm to find the specified number of best vertex assignments between two labeled weighted graphs. The problem is first explicitly formulated as the minimization of a quadratic objective function and then solved by an optimization algorithm based on the recently proposed graduated nonconvexity and concavity procedure (GNCCP). Simulations on both synthetic data and real world images witness the effectiveness of the proposed method.

KW - Feature correspondence

KW - GNCCP

KW - Graph matching

UR - https://www.scopus.com/pages/publications/84922372327

U2 - 10.1016/j.patrec.2014.12.011

DO - 10.1016/j.patrec.2014.12.011

M3 - 文章

AN - SCOPUS:84922372327

SN - 0167-8655

VL - 55

SP - 8

EP - 14

JO - Pattern Recognition Letters

JF - Pattern Recognition Letters

ER -

Outlier robust point correspondence based on GNCCP

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this