EUVIMEDEUVIMED entdecken
Crossref

Dual Uncertainty-Aware Correspondence Adapting and Retaining for Continual Composed Image Retrieval

Haoliang Zhou, Feifei Zhang, Changsheng Xu

IEEE Transactions on Image Processing · 2025 · 34 · 7627-7641

DOI: 10.1109/tip.2025.3628454 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Image Processing, 2025. Verfasst von Haoliang Zhou und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 85 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Tianjin Natural Science Foundation, Beijing Natural Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Recent research in continual learning has primarily focused on unimodal tasks, with limited attention to multimodal tasks such as Composed Image Retrieval (CIR). In this paper, we establish a novel Continual CIR setting named C2IR to simulate the ever-change retrieval demands in the real world. Using the C2IR setting, we identify two significant challenges: intra-task correspondence uncertainty, which hinders the model's ability to manage noisy query-target pair correspondences; and inter-task drift uncertainty, which impedes the model's consistent understanding of relationships, exacerbating catastrophic forgetting across continual tasks. To address these challenges, we propose a Dual Uncertainty-aware Correspondence Adapting and Retaining (U2CAR) framework for C2IR, which leverages uncertainty learning to acquire and consolidate composed correspondence. To ensure reliable composed correspondence inference in each task, we introduce an Uncertainty-based Correspondence Reasoning (UCR) module that estimates and refines the uncertainty in query-target correspondence. Besides, to mitigate catastrophic forgetting of previous tasks, we design an Uncertainty-guided Re-parameterization (URep) paradigm that consolidates valuable composed correspondence knowledge based on the uncertainty variance across various tasks. Extensive experimental results illustrate that our U2CAR significantly outperforms existing methods, demonstrating the robust adaptability and anti-forgetting capabilities of the proposed approach.

Förderung

  • National Natural Science Foundation of China
  • Tianjin Natural Science Foundation
  • Beijing Natural Science Foundation

Zitieren

APA

Zhou, H., Zhang, F., & Xu, C. (2025). Dual Uncertainty-Aware Correspondence Adapting and Retaining for Continual Composed Image Retrieval. IEEE Transactions on Image Processing, 34, 7627-7641. https://doi.org/10.1109/tip.2025.3628454

Vancouver

Zhou H, Zhang F, Xu C. Dual Uncertainty-Aware Correspondence Adapting and Retaining for Continual Composed Image Retrieval. IEEE Transactions on Image Processing. 2025;34:7627-7641. doi:10.1109/tip.2025.3628454

BibTeX
@article{Zhou2025,
  author = {Zhou, Haoliang and Zhang, Feifei and Xu, Changsheng},
  title = {Dual Uncertainty-Aware Correspondence Adapting and Retaining for Continual Composed Image Retrieval},
  journal = {IEEE Transactions on Image Processing},
  year = {2025},
  volume = {34},
  pages = {7627--7641},
  doi = {10.1109/tip.2025.3628454}
}
Kostenlos starten →