Hard Samples Rectification for Unsupervised Cross-domain Person Re-identification

TL;DR

This paper introduces a Hard Samples Rectification (HSR) scheme for unsupervised cross-domain person re-identification, addressing challenges with hard positive and negative samples to improve model performance.

Contribution

The paper proposes a novel HSR learning scheme that rectifies hard positive and negative samples in unsupervised re-ID, enhancing clustering-based methods.

Findings

Achieves promising results on large-scale benchmarks.

Effectively recognizes persons across different views.

Discriminates persons with similar appearances.

Abstract

Person re-identification (re-ID) has received great success with the supervised learning methods. However, the task of unsupervised cross-domain re-ID is still challenging. In this paper, we propose a Hard Samples Rectification (HSR) learning scheme which resolves the weakness of original clustering-based methods being vulnerable to the hard positive and negative samples in the target unlabelled dataset. Our HSR contains two parts, an inter-camera mining method that helps recognize a person under different views (hard positive) and a part-based homogeneity technique that makes the model discriminate different persons but with similar appearance (hard negative). By rectifying those two hard cases, the re-ID model can learn effectively and achieve promising results on two large-scale benchmarks.