MHSA-Net: Multi-Head Self-Attention Network for Occluded Person Re-Identification

TL;DR

This paper introduces MHSA-Net, a novel person re-identification model that uses multi-head self-attention and an attention competition mechanism to focus on key local features and improve performance on occluded and standard re-ID tasks.

Contribution

The paper proposes MHSA-Net with two novel components, MHSAB and ACM, which enhance feature selection and noise filtering for better person re-identification.

Findings

Achieves competitive results on standard and occluded person Re-ID datasets.

Demonstrates that MHSAB and ACM significantly improve re-identification accuracy.

Provides extensive ablation studies confirming the effectiveness of each component.

Abstract

This paper presents a novel person re-identification model, named Multi-Head Self-Attention Network (MHSA-Net), to prune unimportant information and capture key local information from person images. MHSA-Net contains two main novel components: Multi-Head Self-Attention Branch (MHSAB) and Attention Competition Mechanism (ACM). The MHSAB adaptively captures key local person information, and then produces effective diversity embeddings of an image for the person matching. The ACM further helps filter out attention noise and non-key information. Through extensive ablation studies, we verified that the Multi-Head Self-Attention Branch (MHSAB) and Attention Competition Mechanism (ACM) both contribute to the performance improvement of the MHSA-Net. Our MHSA-Net achieves competitive performance in the standard and occluded person Re-ID tasks.