Graph-based Person Signature for Person Re-Identifications

TL;DR

This paper introduces a graph-based method for person re-identification that leverages attributes and visual features, utilizing Graph Convolutional Networks to better capture their correlations and improve matching accuracy.

Contribution

It proposes a novel Graph-based Person Signature framework that integrates attributes and visual features into a graph structure for enhanced person ReID performance.

Findings

Achieves competitive results on Market-1501 and DukeMTMC-ReID datasets.

Outperforms other attribute-based and mask-guided methods.

Effectively models attribute and feature correlations using GCNs.

Abstract

The task of person re-identification (ReID) is to match images of the same person over multiple non-overlapping camera views. Due to the variations in visual factors, previous works have investigated how the person identity, body parts, and attributes benefit the person ReID problem. However, the correlations between attributes, body parts, and within each attribute are not fully utilized. In this paper, we propose a new method to effectively aggregate detailed person descriptions (attributes labels) and visual features (body parts and global features) into a graph, namely Graph-based Person Signature, and utilize Graph Convolutional Networks to learn the topological structure of the visual signature of a person. The graph is integrated into a multi-branch multi-task framework for person re-identification. The extensive experiments are conducted to demonstrate the effectiveness of our proposed approach on two large-scale datasets, including Market-1501 and DukeMTMC-ReID. Our approach achieves competitive results among the state of the art and outperforms other attribute-based or mask-guided methods.