Domain adaptation for person re-identification on new unlabeled data using AlignedReID++

TL;DR

This paper presents a domain adaptation method for person re-identification that leverages AlignedReID++ and pseudo-labels to adapt models trained on one dataset to new, unlabeled data, improving performance without additional annotations.

Contribution

It introduces a domain adaptation workflow using pseudo-labels with AlignedReID++ for person re-identification on unlabeled target data.

Findings

Domain adaptation significantly improves re-identification accuracy.

Unsupervised pseudo-labeling effectively adapts models to new domains.

The method reduces the need for costly data annotation.

Abstract

In the world where big data reigns and there is plenty of hardware prepared to gather a huge amount of non structured data, data acquisition is no longer a problem. Surveillance cameras are ubiquitous and they capture huge numbers of people walking across different scenes. However, extracting value from this data is challenging, specially for tasks that involve human images, such as face recognition and person re-identification. Annotation of this kind of data is a challenging and expensive task. In this work we propose a domain adaptation workflow to allow CNNs that were trained in one domain to be applied to another domain without the need for new annotation of the target data. Our method uses AlignedReID++ as the baseline, trained using a Triplet loss with batch hard. Domain adaptation is done by using pseudo-labels generated using an unsupervised learning strategy. Our results show that domain adaptation techniques really improve the performance of the CNN when applied in the target domain.