Towards Homogeneous Modality Learning and Multi-Granularity Information Exploration for Visible-Infrared Person Re-Identification

TL;DR

This paper proposes a novel approach for visible-infrared person re-identification by using a unified grayscale modality and multi-granularity feature extraction to reduce modality discrepancy and improve retrieval accuracy.

Contribution

It introduces the Aligned Grayscale Modality (AGM) to unify visible and infrared images into a single modality, and employs local feature extraction to enhance cross-modality matching.

Findings

Significantly improves cross-modality retrieval performance.

Reduces modality discrepancy effectively in image and feature space.

Outperforms state-of-the-art methods on benchmark datasets.

Abstract

Visible-infrared person re-identification (VI-ReID) is a challenging and essential task, which aims to retrieve a set of person images over visible and infrared camera views. In order to mitigate the impact of large modality discrepancy existing in heterogeneous images, previous methods attempt to apply generative adversarial network (GAN) to generate the modality-consisitent data. However, due to severe color variations between the visible domain and infrared domain, the generated fake cross-modality samples often fail to possess good qualities to fill the modality gap between synthesized scenarios and target real ones, which leads to sub-optimal feature representations. In this work, we address cross-modality matching problem with Aligned Grayscale Modality (AGM), an unified dark-line spectrum that reformulates visible-infrared dual-mode learning as a gray-gray single-mode learning problem. Specifically, we generate the grasycale modality from the homogeneous visible images. Then, we train a style tranfer model to transfer infrared images into homogeneous grayscale images. In this way, the modality discrepancy is significantly reduced in the image space. In order to reduce the remaining appearance discrepancy, we further introduce a multi-granularity feature extraction network to conduct feature-level alignment. Rather than relying on the global information, we propose to exploit local (head-shoulder) features to assist person Re-ID, which complements each other to form a stronger feature descriptor. Comprehensive experiments implemented on the mainstream evaluation datasets include SYSU-MM01 and RegDB indicate that our method can significantly boost cross-modality retrieval performance against the state of the art methods.