Zero-Assignment Constraint for Graph Matching with Outliers

TL;DR

This paper introduces the zero-assignment constraint (ZAC) to improve graph matching robustness against outliers, providing theoretical insights and an efficient algorithm that outperforms existing methods in accuracy and efficiency.

Contribution

The paper proposes the ZAC method for robust graph matching with outliers, including theoretical analysis and an outlier-robust algorithm that achieves state-of-the-art results.

Findings

Achieves state-of-the-art accuracy in graph matching with outliers.

Significantly reduces incorrect matchings caused by outliers.

Demonstrates superior efficiency compared to existing methods.

Abstract

Graph matching (GM), as a longstanding problem in computer vision and pattern recognition, still suffers from numerous cluttered outliers in practical applications. To address this issue, we present the zero-assignment constraint (ZAC) for approaching the graph matching problem in the presence of outliers. The underlying idea is to suppress the matchings of outliers by assigning zero-valued vectors to the potential outliers in the obtained optimal correspondence matrix. We provide elaborate theoretical analysis to the problem, i.e., GM with ZAC, and figure out that the GM problem with and without outliers are intrinsically different, which enables us to put forward a sufficient condition to construct valid and reasonable objective function. Consequently, we design an efficient outlier-robust algorithm to significantly reduce the incorrect or redundant matchings caused by numerous outliers. Extensive experiments demonstrate that our method can achieve the state-of-the-art performance in terms of accuracy and efficiency, especially in the presence of numerous outliers.