CM-NAS: Cross-Modality Neural Architecture Search for Visible-Infrared Person Re-Identification

TL;DR

This paper introduces CM-NAS, a neural architecture search method focused on optimizing Batch Normalization layer separation to improve cross-modality person re-identification between visible and infrared images, outperforming existing methods.

Contribution

The paper proposes a novel BN-oriented neural architecture search method specifically designed for cross-modality person re-identification tasks.

Findings

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

Achieves 6.70%/6.13% improvement in Rank-1/mAP on SYSU-MM01.

Achieves 12.17%/11.23% improvement in Rank-1/mAP on RegDB.

Abstract

Visible-Infrared person re-identification (VI-ReID) aims to match cross-modality pedestrian images, breaking through the limitation of single-modality person ReID in dark environment. In order to mitigate the impact of large modality discrepancy, existing works manually design various two-stream architectures to separately learn modality-specific and modality-sharable representations. Such a manual design routine, however, highly depends on massive experiments and empirical practice, which is time consuming and labor intensive. In this paper, we systematically study the manually designed architectures, and identify that appropriately separating Batch Normalization (BN) layers is the key to bring a great boost towards cross-modality matching. Based on this observation, the essential objective is to find the optimal separation scheme for each BN layer. To this end, we propose a novel method, named Cross-Modality Neural Architecture Search (CM-NAS). It consists of a BN-oriented search space in which the standard optimization can be fulfilled subject to the cross-modality task. Equipped with the searched architecture, our method outperforms state-of-the-art counterparts in both two benchmarks, improving the Rank-1/mAP by 6.70%/6.13% on SYSU-MM01 and by 12.17%/11.23% on RegDB. Code is released at https://github.com/JDAI-CV/CM-NAS.