Large Sparse Cone Non-negative Matrix Factorization for Image Annotation | |
Tao, Dapeng1; Tao, Dacheng2,3; Li, Xuelong4; Gao, Xinbo5 | |
作者部门 | 光学影像学习与分析中心 |
2017-04-01 | |
发表期刊 | ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY |
ISSN | 2157-6904 |
卷号 | 8期号:3 |
产权排序 | 3 |
摘要 | Image annotation assigns relevant tags to query images based on their semantic contents. Since Non-negative Matrix Factorization (NMF) has the strong ability to learn parts-based representations, recently, a number of algorithms based on NMF have been proposed for image annotation and have achieved good performance. However, most of the efforts have focused on the representations of images and annotations. The properties of the semantic parts have not been well studied. In this article, we revisit the sparseness-constrained NMF (sNMF) proposed by Hoyer [ 2004]. By endowing the sparseness constraint with a geometric interpretation and sNMF with theoretical analyses of the generalization ability, we show that NMF with such a sparseness constraint has three advantages for image annotation tasks: (i) The sparseness constraint is more l(0)-norm oriented than the l(0)-norm-based sparseness, which significantly enhances the ability of NMF to robustly learn semantic parts. (ii) The sparseness constraint has a large cone interpretation and thus allows the reconstruction error of NMF to be smaller, which means that the learned semantic parts are more powerful to represent images for tagging. (iii) The learned semantic parts are less correlated, which increases the discriminative ability for annotating images. Moreover, we present a new efficient large sparse cone NMF (LsCNMF) algorithm to optimize the sNMF problem by employing the Nesterov's optimal gradient method. We conducted experiments on the PASCAL VOC07 dataset and demonstrated the effectiveness of LsCNMF for image annotation. |
文章类型 | Article |
关键词 | Non-negative Matrix Factorization Image Annotation Nesterovs Optimal Gradient Sparseness Constraint |
WOS标题词 | Science & Technology ; Technology |
DOI | 10.1145/2987379 |
收录类别 | SCI ; EI |
关键词[WOS] | SOCIAL MULTIMEDIA ; RECOGNITION ; NETWORKS ; WAVELET |
语种 | 英语 |
WOS研究方向 | Computer Science |
项目资助者 | National Natural Science Foundation of China(61572486 ; Yunnan Natural Science Funds(2016FB105) ; Guangdong Natural Science Funds(2014A030310252) ; Shenzhen Technology Project(JCYJ20140901003939001) ; Opening Project of State Key Laboratory of Digital Publishing Technology ; Program for Excellent Young Talents of Yunnan University ; Australian Research Council(FT-130101457 ; Program for Changjiang Scholars and Innovative Research Team in University of China(IRT13088) ; 61432014) ; DP-140102164 ; LE-140100061) |
WOS类目 | Computer Science, Artificial Intelligence ; Computer Science, Information Systems |
WOS记录号 | WOS:000400160800004 |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://ir.opt.ac.cn/handle/181661/28864 |
专题 | 光谱成像技术研究室 |
作者单位 | 1.Yunnan Univ, Sch Informat Sci & Engn, Kunming 650091, Peoples R China 2.Univ Sydney, Sch Informat Technol, J12-318 Cleveland St, Darlington, NSW 2008, Australia 3.Univ Sydney, Fac Engn & Informat Technol, J12-318 Cleveland St, Darlington, NSW 2008, Australia 4.Chinese Acad Sci, Ctr OPT IMagery Anal & Learning OPTIMAL, State Key Lab Transient Opt & Photon, Xian Inst Opt & Precis Mech, Xian 710119, Peoples R China 5.Xidian Univ, State Key Lab Integrated Serv Networks, Sch Elect Engn, Xian 710071, Peoples R China |
推荐引用方式 GB/T 7714 | Tao, Dapeng,Tao, Dacheng,Li, Xuelong,et al. Large Sparse Cone Non-negative Matrix Factorization for Image Annotation[J]. ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY,2017,8(3). |
APA | Tao, Dapeng,Tao, Dacheng,Li, Xuelong,&Gao, Xinbo.(2017).Large Sparse Cone Non-negative Matrix Factorization for Image Annotation.ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY,8(3). |
MLA | Tao, Dapeng,et al."Large Sparse Cone Non-negative Matrix Factorization for Image Annotation".ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY 8.3(2017). |
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Large Sparse Cone No(711KB) | 期刊论文 | 作者接受稿 | 限制开放 | CC BY-NC-SA | 请求全文 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论