Dynamic Clustering and Cluster Contrastive Learning for Unsupervised Person Re-identification

TL;DR

This paper introduces a novel unsupervised person re-identification method combining dynamic clustering, contrastive learning, and label smoothing to improve feature alignment and model performance.

Contribution

It proposes a dynamic clustering parameter scheduler and a cluster contrastive learning approach that adaptively align features, enhancing unsupervised Re-ID performance.

Findings

Outperforms previous state-of-the-art methods on public datasets.

Effectively balances contrastive and self-supervised learning.

Achieves high computational efficiency.

Abstract

Unsupervised Re-ID methods aim at learning robust and discriminative features from unlabeled data. However, existing methods often ignore the relationship between module parameters of Re-ID framework and feature distributions, which may lead to feature misalignment and hinder the model performance. To address this problem, we propose a dynamic clustering and cluster contrastive learning (DCCC) method. Specifically, we first design a dynamic clustering parameters scheduler (DCPS) which adjust the hyper-parameter of clustering to fit the variation of intra- and inter-class distances. Then, a dynamic cluster contrastive learning (DyCL) method is designed to match the cluster representation vectors' weights with the local feature association. Finally, a label smoothing soft contrastive loss ($L_{ss}$) is built to keep the balance between cluster contrastive learning and self-supervised learning with low computational consumption and high computational efficiency. Experiments on several widely used public datasets validate the effectiveness of our proposed DCCC which outperforms previous state-of-the-art methods by achieving the best performance.