Towards Fully Decoupled End-to-End Person Search

TL;DR

This paper introduces a fully decoupled end-to-end person search model that separately optimizes detection and re-identification tasks, significantly improving performance over previous partially decoupled methods.

Contribution

It proposes a task-incremental network that fully decouples detection and re-id tasks, enabling independent training and better overall person search performance.

Findings

Outperforms existing methods on standard benchmarks.

Demonstrates effective independent learning of detection and re-id.

Achieves higher accuracy in person search tasks.

Abstract

End-to-end person search aims to jointly detect and re-identify a target person in raw scene images with a unified model. The detection task unifies all persons while the re-id task discriminates different identities, resulting in conflict optimal objectives. Existing works proposed to decouple end-to-end person search to alleviate such conflict. Yet these methods are still sub-optimal on one or two of the sub-tasks due to their partially decoupled models, which limits the overall person search performance. In this paper, we propose to fully decouple person search towards optimal person search. A task-incremental person search network is proposed to incrementally construct an end-to-end model for the detection and re-id sub-task, which decouples the model architecture for the two sub-tasks. The proposed task-incremental network allows task-incremental training for the two conflicting tasks. This enables independent learning for different objectives thus fully decoupled the model for persons earch. Comprehensive experimental evaluations demonstrate the effectiveness of the proposed fully decoupled models for end-to-end person search.