Tensorized multi-view subspace representation learning
- Submitting institution
-
University of Greenwich
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 29894
- Type
- D - Journal article
- DOI
-
10.1007/s11263-020-01307-0
- Title of journal
- International Journal of Computer Vision
- Article number
- -
- First page
- 2344
- Volume
- 128
- Issue
- 8-9
- ISSN
- 0920-5691
- Open access status
- Not compliant
- Month of publication
- February
- Year of publication
- 2020
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
5
- Research group(s)
-
-
- Citation count
- 2
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Recently, multi-view data have gained prominence in many real-world applications, as data are increasingly collected from various sources, or represented by multiple features types. Moreover, real-world data are usually associated with priors such as label information, which can improve the discriminability of representation features. To utilise these two cues, this work is the first to propose advanced representation learning by making use of multiple views and prior constraints within a unified framework. The proposed approach achieves more accurate representation and leads to more accurate clustering and classification in comparison to the state-of-the-art such as RMSC (Robust multi-view spectral clustering).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
