Saved in:
Bibliographic Details
Main Authors: Wang, Haiguang, Wu, Yu, Wu, Mengxia, Min, Cao, Zhang, Min
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2406.10845
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914846573854720
author Wang, Haiguang
Wu, Yu
Wu, Mengxia
Min, Cao
Zhang, Min
author_facet Wang, Haiguang
Wu, Yu
Wu, Mengxia
Min, Cao
Zhang, Min
contents Text-based person search aims at retrieving images of a particular person based on a given textual description. A common solution for this task is to directly match the entire images and texts, i.e., global alignment, which fails to deal with discerning specific details that discriminate against appearance-similar people. As a result, some works shift their attention towards local alignment. One group matches fine-grained parts using forward attention weights of the transformer yet underutilizes information. Another implicitly conducts local alignment by reconstructing masked parts based on unmasked context yet with a biased masking strategy. All limit performance improvement. This paper proposes the Local Alignment from Image-Phrase modeling (LAIP) framework, with Bidirectional Attention-weighted local alignment (BidirAtt) and Mask Phrase Modeling (MPM) module.BidirAtt goes beyond the typical forward attention by considering the gradient of the transformer as backward attention, utilizing two-sided information for local alignment. MPM focuses on mask reconstruction within the noun phrase rather than the entire text, ensuring an unbiased masking strategy. Extensive experiments conducted on the CUHK-PEDES, ICFG-PEDES, and RSTPReid datasets demonstrate the superiority of the LAIP framework over existing methods.
format Preprint
id arxiv_https___arxiv_org_abs_2406_10845
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle LAIP: Learning Local Alignment from Image-Phrase Modeling for Text-based Person Search
Wang, Haiguang
Wu, Yu
Wu, Mengxia
Min, Cao
Zhang, Min
Computer Vision and Pattern Recognition
Text-based person search aims at retrieving images of a particular person based on a given textual description. A common solution for this task is to directly match the entire images and texts, i.e., global alignment, which fails to deal with discerning specific details that discriminate against appearance-similar people. As a result, some works shift their attention towards local alignment. One group matches fine-grained parts using forward attention weights of the transformer yet underutilizes information. Another implicitly conducts local alignment by reconstructing masked parts based on unmasked context yet with a biased masking strategy. All limit performance improvement. This paper proposes the Local Alignment from Image-Phrase modeling (LAIP) framework, with Bidirectional Attention-weighted local alignment (BidirAtt) and Mask Phrase Modeling (MPM) module.BidirAtt goes beyond the typical forward attention by considering the gradient of the transformer as backward attention, utilizing two-sided information for local alignment. MPM focuses on mask reconstruction within the noun phrase rather than the entire text, ensuring an unbiased masking strategy. Extensive experiments conducted on the CUHK-PEDES, ICFG-PEDES, and RSTPReid datasets demonstrate the superiority of the LAIP framework over existing methods.
title LAIP: Learning Local Alignment from Image-Phrase Modeling for Text-based Person Search
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2406.10845