CASIA OpenIR  > 智能感知与计算研究中心
Large-Scale Weakly Supervised Object Localization via Latent Category Learning
Wang, Chong1; Huang, Kaiqi1; Ren, Weiqiang1; Zhang, Junge1; Maybank, Steve2
Source PublicationIEEE TRANSACTIONS ON IMAGE PROCESSING
2015-04-01
Volume24Issue:4Pages:1371-1385
SubtypeArticle
AbstractLocalizing objects in cluttered backgrounds is challenging under large-scale weakly supervised conditions. Due to the cluttered image condition, objects usually have large ambiguity with backgrounds. Besides, there is also a lack of effective algorithm for large-scale weakly supervised localization in cluttered backgrounds. However, backgrounds contain useful latent information, e.g., the sky in the aeroplane class. If this latent information can be learned, object-background ambiguity can be largely reduced and background can be suppressed effectively. In this paper, we propose the latent category learning (LCL) in large-scale cluttered conditions. LCL is an unsupervised learning method which requires only image-level class labels. First, we use the latent semantic analysis with semantic object representation to learn the latent categories, which represent objects, object parts or backgrounds. Second, to determine which category contains the target object, we propose a category selection strategy by evaluating each category's discrimination. Finally, we propose the online LCL for use in large-scale conditions. Evaluation on the challenging PASCAL Visual Object Class (VOC) 2007 and the large-scale imagenet large-scale visual recognition challenge 2013 detection data sets shows that the method can improve the annotation precision by 10% over previous methods. More importantly, we achieve the detection precision which outperforms previous results by a large margin and can be competitive to the supervised deformable part model 5.0 baseline on both data sets.
KeywordWeakly Supervised Learning Object Localization Latent Semantic Analysis Large-scale
WOS HeadingsScience & Technology ; Technology
WOS KeywordIMAGE CLASSIFICATION ; NETWORKS
Indexed BySCI
Language英语
WOS Research AreaComputer Science ; Engineering
WOS SubjectComputer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS IDWOS:000351088600004
Citation statistics
Cited Times:22[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/8086
Collection智能感知与计算研究中心
Affiliation1.Chinese Acad Sci, Natl Lab Pattern Recognit, Inst Automat, Beijing 100190, Peoples R China
2.Univ London, Birkbeck Coll, Dept Comp Sci & Informat Syst, London WC1E 7HU, England
Recommended Citation
GB/T 7714
Wang, Chong,Huang, Kaiqi,Ren, Weiqiang,et al. Large-Scale Weakly Supervised Object Localization via Latent Category Learning[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2015,24(4):1371-1385.
APA Wang, Chong,Huang, Kaiqi,Ren, Weiqiang,Zhang, Junge,&Maybank, Steve.(2015).Large-Scale Weakly Supervised Object Localization via Latent Category Learning.IEEE TRANSACTIONS ON IMAGE PROCESSING,24(4),1371-1385.
MLA Wang, Chong,et al."Large-Scale Weakly Supervised Object Localization via Latent Category Learning".IEEE TRANSACTIONS ON IMAGE PROCESSING 24.4(2015):1371-1385.
Files in This Item: Download All
File Name/Size DocType Version Access License
Chong Wang_Large-Sca(5133KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Wang, Chong]'s Articles
[Huang, Kaiqi]'s Articles
[Ren, Weiqiang]'s Articles
Baidu academic
Similar articles in Baidu academic
[Wang, Chong]'s Articles
[Huang, Kaiqi]'s Articles
[Ren, Weiqiang]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Wang, Chong]'s Articles
[Huang, Kaiqi]'s Articles
[Ren, Weiqiang]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Chong Wang_Large-Scale Weakly Supervised Object Localization via Latent Category Learning.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

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