Learning Context-Aware Embedding for Person Search

TL;DR

This paper introduces a novel Attention Context-Aware Embedding (ACAE) for person search that leverages contextual information to improve re-identification accuracy, demonstrating state-of-the-art results.

Contribution

The paper proposes ACAE, a new contextual feature head that enhances person features by modeling relations within and across images, and introduces an Image Memory Bank for training efficiency.

Findings

ACAE significantly improves person search performance.

The method achieves state-of-the-art results on benchmark datasets.

ACAE enhances robustness against illumination, pose, and occlusion variations.

Abstract

Person Search is a relevant task that aims to jointly solve Person Detection and Person Re-identification(re-ID). Though most previous methods focus on learning robust individual features for retrieval, it's still hard to distinguish confusing persons because of illumination, large pose variance, and occlusion. Contextual information is practically available in person search task which benefits searching in terms of reducing confusion. To this end, we present a novel contextual feature head named Attention Context-Aware Embedding(ACAE) which enhances contextual information. ACAE repeatedly reviews the person features within and across images to find similar pedestrian patterns, allowing it to implicitly learn to find possible co-travelers and efficiently model contextual relevant instances' relations. Moreover, we propose Image Memory Bank to improve the training efficiency. Experimentally, ACAE shows extensive promotion when built on different one-step methods. Our overall methods achieve state-of-the-art results compared with previous one-step methods.