CASIA OpenIR  > 类脑智能研究中心
Image-level classification by hierarchical structure learning with visual and semantic similarities
Zhang, Chunjie1,4; Cheng, Jian2,3,4; Tian, Qi5
Source PublicationINFORMATION SCIENCES
2018
Volume422Issue:422Pages:271-281
SubtypeArticle
AbstractImage classification methods often use class-level information without considering the distinctive character of each image. Images of the same class may have varied appearances. Besides, visually similar images may not be semantically correlated. To solve these problems, in this paper, we propose a novel image classification method by automatically learning the image-level hierarchical structure (ILHS) using both visual and semantic similarities. We try to generate new representations by exploring both visual and semantic similarities of images. Images are clustered hierarchically to explore their correlations. We then use them for image representations. The diversity of image classes within each cluster is used to re-weight visual similarities. The re-weighted similarities are aggregated to generate new image representations. We conduct image classification experiments on the Caltech-256 dataset, the PASCAL VOC 2007 dataset and the PASCAL VOC 2012 dataset. Experimental results demonstrate the effectiveness of the proposed method. (C) 2017 Elsevier Inc. All rights reserved.
KeywordImage Classification Hierarchical Structure Learning Image-level Modeling Object Categorization
WOS HeadingsScience & Technology ; Technology
DOI10.1016/j.ins.2017.09.024
WOS KeywordLOW-RANK ; SPARSE DECOMPOSITION ; REPRESENTATION ; PREDICTION ; SPACE
Indexed BySCI
Language英语
Funding OrganizationNational Natural Science Foundation of China(61303154 ; Scientific Research Key Program of Beijing Municipal Commission of Education(KZ201610005012) ; ARO grant(W911NF-15-1-0290) ; NEC Laboratory of America ; NEC Laboratory of Blippar ; National Science Foundation of China (NSFC)(61429201) ; 61332016)
WOS Research AreaComputer Science
WOS SubjectComputer Science, Information Systems
WOS IDWOS:000414887900016
Citation statistics
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/15316
Collection类脑智能研究中心
Affiliation1.Chinese Acad Sci, Inst Automat, Res Ctr Brain Inspired Intelligence, Beijing 100190, Peoples R China
2.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
3.Chinese Acad Sci, Ctr Excellence Brain Sci & Intelligence Technol, Beijing, Peoples R China
4.Univ Chinese Acad Sci, Beijing 100049, Peoples R China
5.Univ Texas San Antonio, Dept Comp Sci, San Antonio, TX 78249 USA
Recommended Citation
GB/T 7714
Zhang, Chunjie,Cheng, Jian,Tian, Qi. Image-level classification by hierarchical structure learning with visual and semantic similarities[J]. INFORMATION SCIENCES,2018,422(422):271-281.
APA Zhang, Chunjie,Cheng, Jian,&Tian, Qi.(2018).Image-level classification by hierarchical structure learning with visual and semantic similarities.INFORMATION SCIENCES,422(422),271-281.
MLA Zhang, Chunjie,et al."Image-level classification by hierarchical structure learning with visual and semantic similarities".INFORMATION SCIENCES 422.422(2018):271-281.
Files in This Item: Download All
File Name/Size DocType Version Access License
1-s2.0-S002002551730(3107KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
ins 录用邮件.pdf(111KB) 开放获取--View Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Zhang, Chunjie]'s Articles
[Cheng, Jian]'s Articles
[Tian, Qi]'s Articles
Baidu academic
Similar articles in Baidu academic
[Zhang, Chunjie]'s Articles
[Cheng, Jian]'s Articles
[Tian, Qi]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Zhang, Chunjie]'s Articles
[Cheng, Jian]'s Articles
[Tian, Qi]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 1-s2.0-S0020025517309568-main.pdf
Format: Adobe PDF
File name: ins 录用邮件.pdf
Format: Adobe PDF
This file does not support browsing at this time
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.