Residential College | false |
Status | 已發表Published |
COMPOSED IMAGE RETRIEVAL WITH TEXT FEEDBACK VIA MULTI-GRAINED UNCERTAINTY REGULARIZATION | |
Chen, Yiyang1,2; Zheng, Zhedong3![]() ![]() | |
2024 | |
Conference Name | 12th International Conference on Learning Representations, ICLR 2024 |
Source Publication | 12th International Conference on Learning Representations, ICLR 2024
![]() |
Pages | 200372 |
Conference Date | 7-11 May 2024 |
Conference Place | Hybrid, Vienna |
Publisher | International Conference on Learning Representations, ICLR |
Abstract | We investigate composed image retrieval with text feedback. Users gradually look for the target of interest by moving from coarse to fine-grained feedback. However, existing methods merely focus on the latter, i.e., fine-grained search, by harnessing positive and negative pairs during training. This pair-based paradigm only considers the one-to-one distance between a pair of specific points, which is not aligned with the one-to-many coarse-grained retrieval process and compromises the recall rate. In an attempt to fill this gap, we introduce a unified learning approach to simultaneously modeling the coarse- and fine-grained retrieval by considering the multi-grained uncertainty. The key idea underpinning the proposed method is to integrate fine- and coarse-grained retrieval as matching data points with small and large fluctuations, respectively. Specifically, our method contains two modules: uncertainty modeling and uncertainty regularization. (1) The uncertainty modeling simulates the multi-grained queries by introducing identically distributed fluctuations in the feature space. (2) Based on the uncertainty modeling, we further introduce uncertainty regularization to adapt the matching objective according to the fluctuation range. Compared with existing methods, the proposed strategy explicitly prevents the model from pushing away potential candidates in the early stage, and thus improves the recall rate. On the three public datasets, i.e., FashionIQ, Fashion200k, and Shoes, the proposed method has achieved +4.03%, +3.38%, and +2.40% Recall@50 accuracy over a strong baseline, respectively. |
DOI | 10.48550/arXiv.2211.07394 |
URL | View the original |
Language | 英語English |
Scopus ID | 2-s2.0-85190545565 |
Fulltext Access | |
Citation statistics | |
Document Type | Conference paper |
Collection | Faculty of Science and Technology INSTITUTE OF COLLABORATIVE INNOVATION DEPARTMENT OF COMPUTER AND INFORMATION SCIENCE |
Corresponding Author | Zheng, Zhedong |
Affiliation | 1.Sea-NExT Joint Lab, National University of Singapore, Singapore 2.Tsinghua University, China 3.Faculty of Science and Technology, Institute of Collaborative Innovation, University of Macau, Macao |
Corresponding Author Affilication | INSTITUTE OF COLLABORATIVE INNOVATION |
Recommended Citation GB/T 7714 | Chen, Yiyang,Zheng, Zhedong,Ji, Wei,et al. COMPOSED IMAGE RETRIEVAL WITH TEXT FEEDBACK VIA MULTI-GRAINED UNCERTAINTY REGULARIZATION[C]:International Conference on Learning Representations, ICLR, 2024, 200372. |
APA | Chen, Yiyang., Zheng, Zhedong., Ji, Wei., Qu, Leigang., & Chua, Tat Seng (2024). COMPOSED IMAGE RETRIEVAL WITH TEXT FEEDBACK VIA MULTI-GRAINED UNCERTAINTY REGULARIZATION. 12th International Conference on Learning Representations, ICLR 2024, 200372. |
Files in This Item: | There are no files associated with this item. |
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment