Leaning Compact and Representative Features for Cross-Modality Person Re-Identification

TL;DR

This paper introduces a new angular triplet loss and a cross-modality knowledge distillation loss to improve cross-modality person re-identification by learning more discriminative and modality-invariant features, achieving superior results.

Contribution

The paper proposes a novel angular triplet loss and a cross-modality knowledge distillation loss to enhance feature discrimination and modality invariance in VI Re-ID.

Findings

Outperforms state-of-the-art methods on RegDB and SYSU-MM01 datasets.

Effectively reduces modality discrepancy with angularly discriminative features.

Demonstrates significant improvement in cross-modality person re-identification accuracy.

Abstract

This paper pays close attention to the cross-modality visible-infrared person re-identification (VI Re-ID) task, which aims to match pedestrian samples between visible and infrared modes. In order to reduce the modality-discrepancy between samples from different cameras, most existing works usually use constraints based on Euclidean metric. Because of the Euclidean based distance metric strategy cannot effectively measure the internal angles between the embedded vectors, the existing solutions cannot learn the angularly discriminative feature embedding. Since the most important factor affecting the classification task based on embedding vector is whether there is an angularly discriminative feature space, in this paper, we present a new loss function called Enumerate Angular Triplet (EAT) loss. Also, motivated by the knowledge distillation, to narrow down the features between different modalities before feature embedding, we further present a novel Cross-Modality Knowledge Distillation (CMKD) loss. Benefit from the above two considerations, the embedded features are discriminative enough in a way to tackle modality-discrepancy problem. The experimental results on RegDB and SYSU-MM01 datasets have demonstrated that the proposed method is superior to the other most advanced methods in terms of impressive performance. Code is available at https://github.com/IVIPLab/LCCRF.