Lifelong Person Search

TL;DR

This paper introduces lifelong person search, an incremental learning framework that maintains performance across multiple datasets by using knowledge distillation and rehearsal-based instance matching, addressing catastrophic forgetting.

Contribution

It proposes a novel end-to-end lifelong person search framework combining knowledge distillation and rehearsal-based instance matching to handle multiple datasets sequentially.

Findings

Significantly improves detection and re-identification performance across datasets.

Effectively mitigates catastrophic forgetting in lifelong person search.

Outperforms existing methods in preserving old domain knowledge.

Abstract

Person search is the task to localize a query person in gallery datasets of scene images. Existing methods have been mainly developed to handle a single target dataset only, however diverse datasets are continuously given in practical applications of person search. In such cases, they suffer from the catastrophic knowledge forgetting in the old datasets when trained on new datasets. In this paper, we first introduce a novel problem of lifelong person search (LPS) where the model is incrementally trained on the new datasets while preserving the knowledge learned in the old datasets. We propose an end-to-end LPS framework that facilitates the knowledge distillation to enforce the consistency learning between the old and new models by utilizing the prototype features of the foreground persons as well as the hard background proposals in the old domains. Moreover, we also devise the rehearsal-based instance matching to further improve the discrimination ability in the old domains by using the unlabeled person instances additionally. Experimental results demonstrate that the proposed method achieves significantly superior performance of both the detection and re-identification to preserve the knowledge learned in the old domains compared with the existing methods.