Complementary Pseudo Labels For Unsupervised Domain Adaptation On Person Re-identification

TL;DR

This paper introduces a joint learning framework for unsupervised domain adaptation in person re-identification, leveraging complementary pseudo labels to improve feature embeddings and achieve state-of-the-art results.

Contribution

It proposes combining high precision neighbor pseudo labels with high recall group pseudo labels, using a similarity-aggregating loss to enhance re-ID performance.

Findings

Achieves state-of-the-art results on three large-scale datasets.

Effectively combines different pseudo label types for better domain adaptation.

Demonstrates robustness across multiple datasets.

Abstract

In recent years, supervised person re-identification (re-ID) models have received increasing studies. However, these models trained on the source domain always suffer dramatic performance drop when tested on an unseen domain. Existing methods are primary to use pseudo labels to alleviate this problem. One of the most successful approaches predicts neighbors of each unlabeled image and then uses them to train the model. Although the predicted neighbors are credible, they always miss some hard positive samples, which may hinder the model from discovering important discriminative information of the unlabeled domain. In this paper, to complement these low recall neighbor pseudo labels, we propose a joint learning framework to learn better feature embeddings via high precision neighbor pseudo labels and high recall group pseudo labels. The group pseudo labels are generated by transitively merging neighbors of different samples into a group to achieve higher recall. However, the merging operation may cause subgroups in the group due to imperfect neighbor predictions. To utilize these group pseudo labels properly, we propose using a similarity-aggregating loss to mitigate the influence of these subgroups by pulling the input sample towards the most similar embeddings. Extensive experiments on three large-scale datasets demonstrate that our method can achieve state-of-the-art performance under the unsupervised domain adaptation re-ID setting.