Ensemble of tensor classifiers based on the higher-order singular value decomposition

Authors:
Bogusław Cyganek
Affiliations:
AGH University of Science and Technology, Kraków, Poland
Venue:
HAIS'12 Proceedings of the 7th international conference on Hybrid Artificial Intelligent Systems - Volume Part II
Year:
2012

Citing 14
Cited 1

On Combining Classifiers

IEEE Transactions on Pattern Analysis and Machine Intelligence
A Multilinear Singular Value Decomposition

SIAM Journal on Matrix Analysis and Applications
A Database for Handwritten Text Recognition Research

IEEE Transactions on Pattern Analysis and Machine Intelligence
Multilinear Analysis of Image Ensembles: TensorFaces

ECCV '02 Proceedings of the 7th European Conference on Computer Vision-Part I
Pattern Classification (2nd Edition)

Pattern Classification (2nd Edition)
Combining Pattern Classifiers: Methods and Algorithms

Combining Pattern Classifiers: Methods and Algorithms
Bagging Equalizes Influence

Machine Learning
An Introduction to 3D Computer Vision Techniques and Algorithms

An Introduction to 3D Computer Vision Techniques and Algorithms
Handwritten digit classification using higher order singular value decomposition

Pattern Recognition
Eigenfaces for recognition

Journal of Cognitive Neuroscience
Pattern Recognition, Fourth Edition

Pattern Recognition, Fourth Edition
Tensor Decompositions and Applications

SIAM Review
Algorithm of designing compound recognition system on the basis of combining classifiers with simultaneous splitting feature space into competence areas

Pattern Analysis & Applications
Computer Vision: Algorithms and Applications

Computer Vision: Algorithms and Applications

Embedding of the extended euclidean distance into pattern recognition with higher-order singular value decomposition of prototype tensors

CISIM'12 Proceedings of the 11th IFIP TC 8 international conference on Computer Information Systems and Industrial Management

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we present an ensemble composed of classifiers operating with multi-dimensional data. Classification is performed in tensor spaces spanned by the basis obtained from the Higher-Order Singular Value Decomposition of the pattern tensors. These showed superior results when processing multi-dimensional data, such as sequences of images. However, multi-dimensionality leads to excessive computational requirements. The proposed method alleviates this problem, first by partitioning the input dataset, and then by feeding each partition into a separate tensor classifiers of the ensemble. Despite the computational advantages, also accuracy of the ensemble showed to be higher compared to a single classifier case. The method was tested in the context of object recognition in computer vision. In the paper we discuss also methods of input image prefiltering in order to increase accuracy. The conducted experiments show high efficacy of the proposed solution.