Contextual Non-Local Alignment over Full-Scale Representation for Text-Based Person Search

TL;DR

This paper introduces NAFS, a novel method for text-based person search that adaptively aligns visual and textual features across all scales using a non-local attention mechanism, significantly improving retrieval accuracy.

Contribution

The paper proposes a full-scale, adaptive alignment approach with a staircase network and locality-constrained BERT, addressing limitations of scale-specific alignment methods.

Findings

Outperforms state-of-the-art by 5.53% top-1 accuracy

Achieves 5.35% improvement in top-5 accuracy

Demonstrates effective multi-scale feature alignment

Abstract

Text-based person search aims at retrieving target person in an image gallery using a descriptive sentence of that person. It is very challenging since modal gap makes effectively extracting discriminative features more difficult. Moreover, the inter-class variance of both pedestrian images and descriptions is small. So comprehensive information is needed to align visual and textual clues across all scales. Most existing methods merely consider the local alignment between images and texts within a single scale (e.g. only global scale or only partial scale) then simply construct alignment at each scale separately. To address this problem, we propose a method that is able to adaptively align image and textual features across all scales, called NAFS (i.e.Non-local Alignment over Full-Scale representations). Firstly, a novel staircase network structure is proposed to extract full-scale image features with better locality. Secondly, a BERT with locality-constrained attention is proposed to obtain representations of descriptions at different scales. Then, instead of separately aligning features at each scale, a novel contextual non-local attention mechanism is applied to simultaneously discover latent alignments across all scales. The experimental results show that our method outperforms the state-of-the-art methods by 5.53% in terms of top-1 and 5.35% in terms of top-5 on text-based person search dataset. The code is available at https://github.com/TencentYoutuResearch/PersonReID-NAFS