m-SNE: Multiview Stochastic Neighbor Embedding

Authors:
Bo Xie; Yang Mu; Dacheng Tao; Kaiqi Huang
Affiliations:
Dept. of Electron. Eng., Chinese Univ. of Hong Kong, Hong Kong, China;-;-;-
Venue:
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Year:
2011

Citing 0
Cited 10

Dimensionality reduction by Mixed Kernel Canonical Correlation Analysis

Pattern Recognition
Discriminative information preservation for face recognition

Neurocomputing
Letters: Gabor face recognition by multi-channel classifier fusion of supervised kernel manifold learning

Neurocomputing
LF-EME: Local features with elastic manifold embedding for human action recognition

Neurocomputing
Joint geometry and variability for image recognition

Neurocomputing
A sparse representation method of bimodal biometrics and palmprint recognition experiments

Neurocomputing
Transductive cartoon retrieval by multiple hypergraph learning

ICONIP'12 Proceedings of the 19th international conference on Neural Information Processing - Volume Part III
Multiview Hessian discriminative sparse coding for image annotation

Computer Vision and Image Understanding
Manifold learning by preserving distance orders

Pattern Recognition Letters
Quality of information-based source assessment and selection

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Dimension reduction has been widely used in real-world applications such as image retrieval and document classification. In many scenarios, different features (or multiview data) can be obtained, and how to duly utilize them is a challenge. It is not appropriate for the conventional concatenating strategy to arrange features of different views into a long vector. That is because each view has its specific statistical property and physical interpretation. Even worse, the performance of the concatenating strategy will deteriorate if some views are corrupted by noise. In this paper, we propose a multiview stochastic neighbor embedding (m-SNE) that systematically integrates heterogeneous features into a unified representation for subsequent processing based on a probabilistic framework. Compared with conventional strategies, our approach can automatically learn a combination coefficient for each view adapted to its contribution to the data embedding. This combination coefficient plays an important role in utilizing the complementary information in multiview data. Also, our algorithm for learning the combination coefficient converges at a rate of O(1/k2), which is the optimal rate for smooth problems. Experiments on synthetic and real data sets suggest the effectiveness and robustness of m-SNE for data visualization, image retrieval, object categorization, and scene recognition.