From Data Deluge to Data Curation: A Filtering-WoRA Paradigm for Efficient Text-based Person Search

Jintao Sun, Hao Fei, Gangyi Ding, Zhedong Zheng*

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Citation (Scopus)

Abstract

In text-based person search endeavors, data generation has emerged as a prevailing practice, addressing concerns over privacy preservation and the arduous task of manual annotation. Although the number of synthesized data can be infinite in theory, the scientific conundrum persists that how much generated data optimally fuels subsequent model training. We observe that only a subset of the data in these constructed datasets plays a decisive role. Therefore, we introduce a new Filtering-WoRA paradigm, which contains a filtering algorithm to identify this crucial data subset and WoRA (Weighted Low-Rank Adaptation) learning strategy for light fine-tuning. The filtering algorithm is based on the cross-modality relevance to remove the lots of coarse matching synthesis pairs. As the number of data decreases, we do not need to fine-tune the entire model. Therefore, we propose a WoRA learning strategy to efficiently update a minimal portion of model parameters. WoRA streamlines the learning process, enabling heightened efficiency in extracting knowledge from fewer, yet potent, data instances. Extensive experimentation validates the efficacy of pretraining, where our model achieves advanced and efficient retrieval performance on challenging real-world benchmarks. Notably, on the CUHK-PEDES dataset, we have achieved a competitive mAP of 67.02% while reducing model training time by 19.82%.

Original languageEnglish
Title of host publicationWWW 2025 - Proceedings of the ACM Web Conference
PublisherAssociation for Computing Machinery, Inc
Pages2341-2351
Number of pages11
ISBN (Electronic)9798400712746
DOIs
Publication statusPublished - 28 Apr 2025
Externally publishedYes
Event34th ACM Web Conference, WWW 2025 - Sydney, Australia
Duration: 28 Apr 20252 May 2025

Publication series

NameWWW 2025 - Proceedings of the ACM Web Conference

Conference

Conference34th ACM Web Conference, WWW 2025
Country/TerritoryAustralia
CitySydney
Period28/04/252/05/25

Keywords

  • Data-centric Learning
  • Low-Rank Adaptation
  • Text-based Person Search
  • Visual-language Pre-training

Cite this