Self-similarity Grouping: A Simple Unsupervised Cross Domain Adaptation Approach for Person Re-identification

TL;DR

This paper introduces Self-similarity Grouping (SSG), an unsupervised method for person re-identification that leverages natural sample similarities to build pseudo labels, significantly improving cross-domain re-ID performance.

Contribution

The paper proposes a novel unsupervised clustering-based approach, SSG, for domain adaptation in person re-ID, and extends it with SSG++ for one-shot open-set adaptation.

Findings

SSG outperforms state-of-the-art by over 4.6% in mAP.

SSG++ further improves mAP by up to 10.7%.

Method achieves stable training through iterative grouping and learning.

Abstract

Domain adaptation in person re-identification (re-ID) has always been a challenging task. In this work, we explore how to harness the natural similar characteristics existing in the samples from the target domain for learning to conduct person re-ID in an unsupervised manner. Concretely, we propose a Self-similarity Grouping (SSG) approach, which exploits the potential similarity (from global body to local parts) of unlabeled samples to automatically build multiple clusters from different views. These independent clusters are then assigned with labels, which serve as the pseudo identities to supervise the training process. We repeatedly and alternatively conduct such a grouping and training process until the model is stable. Despite the apparent simplify, our SSG outperforms the state-of-the-arts by more than 4.6% (DukeMTMC to Market1501) and 4.4% (Market1501 to DukeMTMC) in mAP, respectively. Upon our SSG, we further introduce a clustering-guided semisupervised approach named SSG ++ to conduct the one-shot domain adaption in an open set setting (i.e. the number of independent identities from the target domain is unknown). Without spending much effort on labeling, our SSG ++ can further promote the mAP upon SSG by 10.7% and 6.9%, respectively. Our Code is available at: https://github.com/OasisYang/SSG .