Unsupervised Clustering Active Learning for Person Re-identification

TL;DR

This paper introduces UCAL, an active learning approach that combines unsupervised person re-identification with minimal human annotations to improve model performance efficiently.

Contribution

It proposes a novel incremental clustering method that selects representative pairs for annotation, reducing human effort while enhancing unsupervised re-id models.

Findings

Outperforms state-of-the-art active learning methods on benchmark datasets.

Requires minimal human annotation effort due to centroid-pair selection.

Achieves competitive re-id accuracy with limited labeled data.

Abstract

Supervised person re-identification (re-id) approaches require a large amount of pairwise manual labeled data, which is not applicable in most real-world scenarios for re-id deployment. On the other hand, unsupervised re-id methods rely on unlabeled data to train models but performs poorly compared with supervised re-id methods. In this work, we aim to combine unsupervised re-id learning with a small number of human annotations to achieve a competitive performance. Towards this goal, we present a Unsupervised Clustering Active Learning (UCAL) re-id deep learning approach. It is capable of incrementally discovering the representative centroid-pairs and requiring human annotate them. These few labeled representative pairwise data can improve the unsupervised representation learning model with other large amounts of unlabeled data. More importantly, because the representative centroid-pairs are selected for annotation, UCAL can work with very low-cost human effort. Extensive experiments demonstrate the superiority of the proposed model over state-of-the-art active learning methods on three re-id benchmark datasets.