Domain Adaptive Person Search

TL;DR

This paper introduces Domain Adaptive Person Search (DAPS), a framework that enhances the generalization of person search models across different domains by domain alignment and pseudo-labeling, outperforming existing methods.

Contribution

The paper proposes a novel domain adaptive framework with image and instance-level alignment and a dynamic clustering strategy for unlabeled target data.

Findings

Achieves 34.7% mAP and 80.6% top-1 on PRW dataset.

Surpasses baseline by a large margin and even outperforms some supervised methods.

Effective domain adaptation for person search tasks.

Abstract

Person search is a challenging task which aims to achieve joint pedestrian detection and person re-identification (ReID). Previous works have made significant advances under fully and weakly supervised settings. However, existing methods ignore the generalization ability of the person search models. In this paper, we take a further step and present Domain Adaptive Person Search (DAPS), which aims to generalize the model from a labeled source domain to the unlabeled target domain. Two major challenges arises under this new setting: one is how to simultaneously solve the domain misalignment issue for both detection and Re-ID tasks, and the other is how to train the ReID subtask without reliable detection results on the target domain. To address these challenges, we propose a strong baseline framework with two dedicated designs. 1) We design a domain alignment module including image-level and task-sensitive instance-level alignments, to minimize the domain discrepancy. 2) We take full advantage of the unlabeled data with a dynamic clustering strategy, and employ pseudo bounding boxes to support ReID and detection training on the target domain. With the above designs, our framework achieves 34.7% in mAP and 80.6% in top-1 on PRW dataset, surpassing the direct transferring baseline by a large margin. Surprisingly, the performance of our unsupervised DAPS model even surpasses some of the fully and weakly supervised methods. The code is available at https://github.com/caposerenity/DAPS.