Structure Preserving Multi-View Dimensionality Reduction

Zhan Wang, Lizhi Wang, Hua Huang

Abstract

The multi-view features from multimedia data in the real-world are usually high-dimensional. How to simultaneously reduce their dimensions and explore the complementary information among multi-view features is of vital importance but challenging. In this paper, we propose a novel unsupervised method named structure preserving multi-view dimensionality reduction (SPMDR). We first propose a bilinear low-rank representation with an orthogonal constraint in the learning subspace. Then, we construct a 3-order rotated tensor among the low-rank coefficient matrices and utilize tensor nuclear norm to capture complementary information among multi-view representations. Finally, we develop a numerical algorithm for solving the proposed model. Our method is robust to noisy data and can capture the complex correlations among multi-view features. Experimental results on recognition tasks demonstrate the superior performance of SPMDR.
BibTeX
@inproceedings{wang2020structure,
  author = {Zhan Wang and Lizhi Wang and Hua Huang},
  title = {{Structure Preserving Multi-View Dimensionality Reduction}},
  booktitle = {ICME},
  year = {2020},
  doi = {10.1109/icme46284.2020.9102900},
}

Loading related content...

Figure