Cross-View Feature Learning via Structures Unlocking Based on Robust Low-Rank Constraint
The cross-view multimedia are widely existed and attract many attentions in recent years. Nevertheless, it is noted that the phenomenon, that data in different classes from same view are more similar than that in same class from different views, is usually presented for cross-view multimedia data. T...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9025222/ |
| _version_ | 1811273947907358720 |
|---|---|
| author | Ao Li Yu Ding Deyun Chen Guanglu Sun Hailong Jiang Qidi Wu |
| author_facet | Ao Li Yu Ding Deyun Chen Guanglu Sun Hailong Jiang Qidi Wu |
| author_sort | Ao Li |
| collection | DOAJ |
| description | The cross-view multimedia are widely existed and attract many attentions in recent years. Nevertheless, it is noted that the phenomenon, that data in different classes from same view are more similar than that in same class from different views, is usually presented for cross-view multimedia data. The intrinsic imperfection leads disappointing performance for cross-view multimedia recognition or classification. To solve this problem, in this paper, we propose a novel discriminative learning framework with low-rank constraint, which can be applied for view-invariant low-dimensional subspace learning. The advantages of our framework include three aspects. Firstly, to unlock the latent class structure and view structure, a self-expressed model by dual low-rank constraints are presented, which can separate the two manifold structures in the learned subspace. Secondly, two effective discriminative graphs are constructed to guide the affinity relationship of data in the above two low-dimensional projected subspaces respectively. Finally, the joint semantic consensus constraint is designed to be integrated into the learning framework, which can explore the shared and view-specific information for enforcing the view-invariant character in semantic space. Experimental results on several public cross-view multimedia datasets demonstrate that our proposed method outperforms existing excellent subspace learning approaches. |
| first_indexed | 2024-04-12T23:10:04Z |
| format | Article |
| id | doaj.art-004376321f6d478eb9e9e8eb2a99aa48 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-12T23:10:04Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-004376321f6d478eb9e9e8eb2a99aa482022-12-22T03:12:49ZengIEEEIEEE Access2169-35362020-01-018468514686010.1109/ACCESS.2020.29785489025222Cross-View Feature Learning via Structures Unlocking Based on Robust Low-Rank ConstraintAo Li0https://orcid.org/0000-0003-0735-2917Yu Ding1https://orcid.org/0000-0002-5605-6738Deyun Chen2https://orcid.org/0000-0002-5176-7725Guanglu Sun3https://orcid.org/0000-0003-2589-1164Hailong Jiang4https://orcid.org/0000-0003-3246-7177Qidi Wu5https://orcid.org/0000-0003-1922-7435School of Computer Science and Technology, Harbin University of Science and Technology, Harbin, ChinaSchool of Computer Science and Technology, Harbin University of Science and Technology, Harbin, ChinaSchool of Computer Science and Technology, Harbin University of Science and Technology, Harbin, ChinaSchool of Computer Science and Technology, Harbin University of Science and Technology, Harbin, ChinaDepartment of Computer Science, Kent State University, Kent, OH, USACollege of Information and Communication Engineering, Harbin Engineering University, Harbin, ChinaThe cross-view multimedia are widely existed and attract many attentions in recent years. Nevertheless, it is noted that the phenomenon, that data in different classes from same view are more similar than that in same class from different views, is usually presented for cross-view multimedia data. The intrinsic imperfection leads disappointing performance for cross-view multimedia recognition or classification. To solve this problem, in this paper, we propose a novel discriminative learning framework with low-rank constraint, which can be applied for view-invariant low-dimensional subspace learning. The advantages of our framework include three aspects. Firstly, to unlock the latent class structure and view structure, a self-expressed model by dual low-rank constraints are presented, which can separate the two manifold structures in the learned subspace. Secondly, two effective discriminative graphs are constructed to guide the affinity relationship of data in the above two low-dimensional projected subspaces respectively. Finally, the joint semantic consensus constraint is designed to be integrated into the learning framework, which can explore the shared and view-specific information for enforcing the view-invariant character in semantic space. Experimental results on several public cross-view multimedia datasets demonstrate that our proposed method outperforms existing excellent subspace learning approaches.https://ieeexplore.ieee.org/document/9025222/Multimedia analysismulti-view discriminative analysiscross-view feature learninglow-rank representation |
| spellingShingle | Ao Li Yu Ding Deyun Chen Guanglu Sun Hailong Jiang Qidi Wu Cross-View Feature Learning via Structures Unlocking Based on Robust Low-Rank Constraint IEEE Access Multimedia analysis multi-view discriminative analysis cross-view feature learning low-rank representation |
| title | Cross-View Feature Learning via Structures Unlocking Based on Robust Low-Rank Constraint |
| title_full | Cross-View Feature Learning via Structures Unlocking Based on Robust Low-Rank Constraint |
| title_fullStr | Cross-View Feature Learning via Structures Unlocking Based on Robust Low-Rank Constraint |
| title_full_unstemmed | Cross-View Feature Learning via Structures Unlocking Based on Robust Low-Rank Constraint |
| title_short | Cross-View Feature Learning via Structures Unlocking Based on Robust Low-Rank Constraint |
| title_sort | cross view feature learning via structures unlocking based on robust low rank constraint |
| topic | Multimedia analysis multi-view discriminative analysis cross-view feature learning low-rank representation |
| url | https://ieeexplore.ieee.org/document/9025222/ |
| work_keys_str_mv | AT aoli crossviewfeaturelearningviastructuresunlockingbasedonrobustlowrankconstraint AT yuding crossviewfeaturelearningviastructuresunlockingbasedonrobustlowrankconstraint AT deyunchen crossviewfeaturelearningviastructuresunlockingbasedonrobustlowrankconstraint AT guanglusun crossviewfeaturelearningviastructuresunlockingbasedonrobustlowrankconstraint AT hailongjiang crossviewfeaturelearningviastructuresunlockingbasedonrobustlowrankconstraint AT qidiwu crossviewfeaturelearningviastructuresunlockingbasedonrobustlowrankconstraint |