Other Tokens Matter: Exploring Global and Local Features of Vision Transformers for Object Re-Identification

TL;DR

This paper investigates the roles of global and local features in Vision Transformers for object Re-Identification, proposing a novel model that effectively combines these features to improve re-identification accuracy.

Contribution

The paper introduces a Global-Local Transformer (GLTrans) with a Global Aggregation Encoder and Local Multi-layer Fusion to enhance feature representation for object Re-ID.

Findings

Achieves superior performance on four Re-ID benchmarks.

Global and local features mutually enhance each other.

Features from last Transformer layers are highly representative.

Abstract

Object Re-Identification (Re-ID) aims to identify and retrieve specific objects from images captured at different places and times. Recently, object Re-ID has achieved great success with the advances of Vision Transformers (ViT). However, the effects of the global-local relation have not been fully explored in Transformers for object Re-ID. In this work, we first explore the influence of global and local features of ViT and then further propose a novel Global-Local Transformer (GLTrans) for high-performance object Re-ID. We find that the features from last few layers of ViT already have a strong representational ability, and the global and local information can mutually enhance each other. Based on this fact, we propose a Global Aggregation Encoder (GAE) to utilize the class tokens of the last few Transformer layers and learn comprehensive global features effectively. Meanwhile, we propose the Local Multi-layer Fusion (LMF) which leverages both the global cues from GAE and multi-layer patch tokens to explore the discriminative local representations. Extensive experiments demonstrate that our proposed method achieves superior performance on four object Re-ID benchmarks.