iCmSC: Incomplete Cross-Modal Subspace Clustering | |
Wang QQ(王倩倩)1,2; Lian, Huanhuan1; Sun G(孙干)3![]() | |
Department | 机器人学研究室 |
Source Publication | IEEE TRANSACTIONS ON IMAGE PROCESSING
![]() |
ISSN | 1057-7149 |
2021 | |
Volume | 30Pages:305-317 |
Indexed By | SCI |
WOS ID | WOS:000595466700003 |
Contribution Rank | 3 |
Funding Organization | National Natural Science Foundation of ChinaNational Natural Science Foundation of China (NSFC) [61773302, 61906141] ; China Postdoctoral Science FoundationChina Postdoctoral Science Foundation [2019M653564, 2019M663642] ; National Natural Science Foundation of Shaanxi Province [2020JZ-19, 2020JQ-317, 2020JQ-327] ; Innovation Fund of Xidian University ; Initiative Postdocs Supporting Program [BX20190262] |
Keyword | Cross-modal subspace clustering incomplete data deep canonical correlation analysis l(1,2)-norm |
Abstract | Cross-modal clustering aims to cluster the high-similar cross-modal data into one group while separating the dissimilar data. Despite the promising cross-modal methods have developed in recent years, existing state-of-the-arts cannot effectively capture the correlations between cross-modal data when encountering with incomplete cross-modal data, which can gravely degrade the clustering performance. To well tackle the above scenario, we propose a novel incomplete cross-modal clustering method that integrates canonical correlation analysis and exclusive representation, named incomplete Cross-modal Subspace Clustering (i.e., iCmSC). To learn a consistent subspace representation among incomplete cross-modal data, we maximize the intrinsic correlations among different modalities by deep canonical correlation analysis (DCCA), while an exclusive self-expression layer is proposed after the output layers of DCCA. We exploit a l(1,2)-norm regularization in the learned subspace to make the learned representation more discriminative, which makes samples between different clusters mutually exclusive and samples among the same cluster attractive to each other. Meanwhile, the decoding networks are employed to reconstruct the feature representation, and further preserve the structural information among the original cross-modal data. To the end, we demonstrate the effectiveness of the proposed iCmSC via extensive experiments, which can justify that iCmSC achieves consistently large improvement compared with the state-of-thearts. |
Language | 英语 |
Citation statistics | |
Document Type | 期刊论文 |
Identifier | http://ir.sia.cn/handle/173321/28037 |
Collection | 机器人学研究室 |
Corresponding Author | Gao QX(高全学) |
Affiliation | 1.State Key Laboratory of Integrated Services Networks, Xidian University, Xi’an 710071, China 2.Key Lab of Ministry of Education of Intellisense and Image Understanding, School of Telecommunication Engineering, Xidian University, Xi’an 710071, China 3.State Key Laboratory of Robotics, Shenyang Institute of Automation, Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang 110016, China 4.Key Lab of Ministry of Education of Intellisense and Image Understanding, School of Artificial Intelligence, Xidian University,Xi’an 710071, China |
Recommended Citation GB/T 7714 | Wang QQ,Lian, Huanhuan,Sun G,et al. iCmSC: Incomplete Cross-Modal Subspace Clustering[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2021,30:305-317. |
APA | Wang QQ,Lian, Huanhuan,Sun G,Gao QX,&Jiao, Licheng.(2021).iCmSC: Incomplete Cross-Modal Subspace Clustering.IEEE TRANSACTIONS ON IMAGE PROCESSING,30,305-317. |
MLA | Wang QQ,et al."iCmSC: Incomplete Cross-Modal Subspace Clustering".IEEE TRANSACTIONS ON IMAGE PROCESSING 30(2021):305-317. |
Files in This Item: | ||||||
File Name/Size | DocType | Version | Access | License | ||
iCmSC_ Incomplete Cr(6218KB) | 期刊论文 | 出版稿 | 开放获取 | CC BY-NC-SA | View Application Full Text |
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment