A unified formulation of a class of graph matching techniques. (November 2019)

• Record Type: Journal Article
• Authors: Zhu, Yuan; Zhou, Jiufeng; Yan, Hong
• Is Part Of: Pattern recognition. Volume 95(2019:Nov.)
• Page Start: 223
• Page End: 234
• Publication Date: 2019-11
• Subjects: Graph matching -- Relaxation labeling -- Spectral graph theory -- Tensor theory; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2019.06.008
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: We show that graph matching methods based on relaxation labeling, spectral graph theory and the tensor theory have the same mathematical form by employing power iteration technique. We demonstrate that differences among these methods have little impact on the final matching result according to our experiment results. We propose a fast framework to accelerate the matching procedure, which could greatly improve the matching performance. Abstract: In this paper, we show that graph matching methods based on relaxation labeling, spectral graph theory and tensor theory have the same mathematical form by employing power iteration technique. Besides, the differences among these methods are also fully discussed and can be proven that distinctions have little impact on the final matching result. Moreover, we propose a fast compatibility building procedure to accelerate the preprocessing speed which is considered to be the main time consuming part of graph matching. Finally, several experiments are conducted to verify our findings.