Cross Domain Knowledge Transfer for Unsupervised Vehicle Re-identification

TL;DR

This paper introduces a domain adaptation framework for vehicle re-identification that uses image translation and attention-based feature learning to improve performance across different datasets without labeled target data.

Contribution

It proposes VTGAN for style transfer and ATTNet for enhanced feature learning, addressing domain bias in unsupervised vehicle reID.

Findings

Achieves state-of-the-art performance on VehicleID dataset.

Effectively reduces domain bias in vehicle reID.

Improves reID accuracy across different camera views.

Abstract

Vehicle re-identification (reID) is to identify a target vehicle in different cameras with non-overlapping views. When deploy the well-trained model to a new dataset directly, there is a severe performance drop because of differences among datasets named domain bias. To address this problem, this paper proposes an domain adaptation framework which contains an image-to-image translation network named vehicle transfer generative adversarial network (VTGAN) and an attention-based feature learning network (ATTNet). VTGAN could make images from the source domain (well-labeled) have the style of target domain (unlabeled) and preserve identity information of source domain. To further improve the domain adaptation ability for various backgrounds, ATTNet is proposed to train generated images with the attention structure for vehicle reID. Comprehensive experimental results clearly demonstrate that our method achieves excellent performance on VehicleID dataset.