Independent component analysis with graphical correlation: Applications to multi-vision coding

Ryota Yokote; Toshikazu Nakamura; Yasuo Matsuyama

doi:10.1109/IJCNN.2011.6033290

Independent component analysis with graphical correlation: Applications to multi-vision coding

Ryota Yokote^*, Toshikazu Nakamura, Yasuo Matsuyama

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

New algorithms for joint learning of independent component analysis and graphical high-order correlation (GC-ICA: Graphically Correlated ICA) are presented. The presented method has a fixed point style or of the FastICA, however, it comprises independent but correlated subparts. Correlations by teacher signals are also allowed. In spite of such inclusion of the dependency, the presented algorithm shows fast convergence. The converged set of bases has reduced indeterminacy on the ordering. This is equivalent to a self-organization of bases. This method can be used to analyze multiple images simultaneously. Examples are given on images from 3D- stereo videos shots. The correlation of bases on left and right eye views is shown for the first time here. Further speedup using the strategy of the RapidICA is possible.

Original language	English
Title of host publication	Proceedings of the International Joint Conference on Neural Networks
Pages	701-708
Number of pages	8
DOIs	https://doi.org/10.1109/IJCNN.2011.6033290
Publication status	Published - 2011
Event	2011 International Joint Conference on Neural Network, IJCNN 2011 - San Jose, CA Duration: 2011 Jul 31 → 2011 Aug 5

Other

Other	2011 International Joint Conference on Neural Network, IJCNN 2011
City	San Jose, CA
Period	11/7/31 → 11/8/5

ASJC Scopus subject areas

Software
Artificial Intelligence

Access to Document

10.1109/IJCNN.2011.6033290

Cite this

@inproceedings{58b219b26344418ab14da1d4596b63c7,

title = "Independent component analysis with graphical correlation: Applications to multi-vision coding",

abstract = "New algorithms for joint learning of independent component analysis and graphical high-order correlation (GC-ICA: Graphically Correlated ICA) are presented. The presented method has a fixed point style or of the FastICA, however, it comprises independent but correlated subparts. Correlations by teacher signals are also allowed. In spite of such inclusion of the dependency, the presented algorithm shows fast convergence. The converged set of bases has reduced indeterminacy on the ordering. This is equivalent to a self-organization of bases. This method can be used to analyze multiple images simultaneously. Examples are given on images from 3D- stereo videos shots. The correlation of bases on left and right eye views is shown for the first time here. Further speedup using the strategy of the RapidICA is possible.",

author = "Ryota Yokote and Toshikazu Nakamura and Yasuo Matsuyama",

year = "2011",

doi = "10.1109/IJCNN.2011.6033290",

language = "English",

isbn = "9781457710865",

pages = "701--708",

booktitle = "Proceedings of the International Joint Conference on Neural Networks",

note = "2011 International Joint Conference on Neural Network, IJCNN 2011 ; Conference date: 31-07-2011 Through 05-08-2011",

}

TY - GEN

T1 - Independent component analysis with graphical correlation

T2 - 2011 International Joint Conference on Neural Network, IJCNN 2011

AU - Yokote, Ryota

AU - Nakamura, Toshikazu

AU - Matsuyama, Yasuo

PY - 2011

Y1 - 2011

N2 - New algorithms for joint learning of independent component analysis and graphical high-order correlation (GC-ICA: Graphically Correlated ICA) are presented. The presented method has a fixed point style or of the FastICA, however, it comprises independent but correlated subparts. Correlations by teacher signals are also allowed. In spite of such inclusion of the dependency, the presented algorithm shows fast convergence. The converged set of bases has reduced indeterminacy on the ordering. This is equivalent to a self-organization of bases. This method can be used to analyze multiple images simultaneously. Examples are given on images from 3D- stereo videos shots. The correlation of bases on left and right eye views is shown for the first time here. Further speedup using the strategy of the RapidICA is possible.

AB - New algorithms for joint learning of independent component analysis and graphical high-order correlation (GC-ICA: Graphically Correlated ICA) are presented. The presented method has a fixed point style or of the FastICA, however, it comprises independent but correlated subparts. Correlations by teacher signals are also allowed. In spite of such inclusion of the dependency, the presented algorithm shows fast convergence. The converged set of bases has reduced indeterminacy on the ordering. This is equivalent to a self-organization of bases. This method can be used to analyze multiple images simultaneously. Examples are given on images from 3D- stereo videos shots. The correlation of bases on left and right eye views is shown for the first time here. Further speedup using the strategy of the RapidICA is possible.

UR - http://www.scopus.com/inward/record.url?scp=80054741514&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80054741514&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2011.6033290

DO - 10.1109/IJCNN.2011.6033290

M3 - Conference contribution

AN - SCOPUS:80054741514

SN - 9781457710865

SP - 701

EP - 708

BT - Proceedings of the International Joint Conference on Neural Networks

Y2 - 31 July 2011 through 5 August 2011

ER -

Independent component analysis with graphical correlation: Applications to multi-vision coding

Abstract

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this