NL repository
menu
검색
Library
Browse
Communities & Collections
By Date
Authors
Titles
Subject
My Repository
My Account
Receive email updates
Edit Profile
DSpace at EWHA
공과대학
전자전기공학전공
Journal papers
View : 65 Download: 0
Bimodal semantic fusion prototypical network for few-shot classification
Title
Bimodal semantic fusion prototypical network for few-shot classification
Authors
Huang
;
Xilang
;
Choi
;
Seon Han
Ewha Authors
최선한
SCOPUS Author ID
최선한
Issue Date
2024
Journal Title
Information Fusion
ISSN
1566-2535
Citation
Information Fusion vol. 109
Keywords
Feature aggregation
;
Few-shot classification
;
Local and global attention
;
Multi-source information fusion
Publisher
Elsevier B.V.
Indexed
SCIE; SCOPUS
Document Type
Article
Abstract
Few-shot classification learns from a small number of image samples to recognize unseen images. Recent few-shot learning exploits auxiliary text information, such as class labels and names, to obtain more discriminative class prototypes. However, most existing approaches rarely consider using text information as a clue to highlight important feature regions and do not consider feature alignment between prototypes and targets, leading to prototype ambiguity owing to information gaps. To address this issue, a prototype generator module was developed to perform interactions between the text knowledge of the class name and visual feature maps in the spatial and channel dimensions. This module learns how to assign mixture weights to essential regions of each sample feature to obtain informative prototypes. In addition, a feature refinement module was proposed to embed text information into query images without knowing their labels. It generates attention from concatenated features between query and text features through pairwise distance loss. To improve the alignment between the prototype and relevant targets, a prototype calibration module was designed to preserve the important features of the prototype by considering the interrelationships between the prototype and query features. Extensive experiments were conducted on five few-shot classification benchmarks, and the results demonstrated the superiority of the proposed method over state-of-the-art methods in 1-shot and 5-shot settings. © 2024 Elsevier B.V.
DOI
10.1016/j.inffus.2024.102421
Appears in Collections:
공과대학
>
전자전기공학전공
>
Journal papers
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML
Show full item record
Find@EWHA
트윗하기
BROWSE
Communities & Collections
By Date
Authors
Titles
Subject