Robust Graph Structure Learning via Multiple Statistical Tests

TL;DR

This paper introduces a robust method for graph structure learning in computer vision by employing multiple statistical tests to improve the reliability of image similarity-based graph construction, validated through theoretical analysis and empirical experiments.

Contribution

It proposes a novel multiple statistical tests framework for more reliable graph structure learning, along with an efficient matrix form called B-Attention, applicable to image clustering and ReID tasks.

Findings

Enhanced robustness in graph construction demonstrated on benchmark datasets.

Theoretical validation of multiple tests improving reliability.

Empirical results show improved clustering and ReID performance.

Abstract

Graph structure learning aims to learn connectivity in a graph from data. It is particularly important for many computer vision related tasks since no explicit graph structure is available for images for most cases. A natural way to construct a graph among images is to treat each image as a node and assign pairwise image similarities as weights to corresponding edges. It is well known that pairwise similarities between images are sensitive to the noise in feature representations, leading to unreliable graph structures. We address this problem from the viewpoint of statistical tests. By viewing the feature vector of each node as an independent sample, the decision of whether creating an edge between two nodes based on their similarity in feature representation can be thought as a ${\it single}$ statistical test. To improve the robustness in the decision of creating an edge, multiple samples are drawn and integrated by ${\it multiple}$ statistical tests to generate a more reliable similarity measure, consequentially more reliable graph structure. The corresponding elegant matrix form named $\mathcal{B}\textbf{-Attention}$ is designed for efficiency. The effectiveness of multiple tests for graph structure learning is verified both theoretically and empirically on multiple clustering and ReID benchmark datasets. Source codes are available at https://github.com/Thomas-wyh/B-Attention.