Learning Concordant Attention via Target-aware Alignment for Visible-Infrared Person Re-identification

Owing to the large distribution gap between the heterogeneous data in Visible-Infrared Person Re-identification (VI Re-ID), we point out that existing paradigms often suffer from the inter-modal semantic misalignment issue and thus fail to align and compare local details properly. In this paper, we present Concordant Attention Learning (CAL), a novel framework that learns semantic-aligned representations for VI Re-ID. Specifically, we design the Target-aware Concordant Alignment paradigm, which allows target-aware attention adaptation when aligning heterogeneous samples (i.e., adaptive attention adjustment according to the target image being aligned). This is achieved by exploiting the discriminative clues from the modality counterpart and designing effective modality-agnostic correspondence searching strategies. To ensure semantic concordance during the cross-modal retrieval stage, we further propose MatchDistill, which matches the attention patterns across modalities and learns their underlying semantic correlations by bipartite-graph-based similarity modeling and cross-modal knowledge exchange. Extensive experiments on VI Re-ID benchmark datasets demonstrate the effectiveness and superiority of the proposed CAL.

1. Introduction

Person Re-identification (Re-ID) aims to associate person identities across non-overlapping cameras. It has gained increasing attention in recent years due to its practical applications in real-world surveillance systems. Conventional person Re-ID methods mainly focus on retrieving the same identity across visible (RGB) cameras [48], [29], [16], [23], [10]. Despite their remarkable success, they have limited applicability since visible cameras cannot capture discriminative information under poor-lighting conditions (e.g., at night). To improve the illumination robustness, infrared (IR) cameras are widely applied to cooperate with visible ones in real-world surveillance systems. This increases the need to explore the Visible-Infrared Person Re-identification (VI Re-ID) problem, which aims to associate the person images taken by different spectrum cameras to achieve long-term person tracking in 24-hour surveillance systems.