Independent component analysis with graphical correlation

Applications to multi-vision coding

Ryota Yokote, Toshikazu Nakamura, Yasuo Matsuyama

    Research output: Chapter in Book/Report/Conference proceedingConference 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 languageEnglish
    Title of host publicationProceedings of the International Joint Conference on Neural Networks
    Pages701-708
    Number of pages8
    DOIs
    Publication statusPublished - 2011
    Event2011 International Joint Conference on Neural Network, IJCNN 2011 - San Jose, CA
    Duration: 2011 Jul 312011 Aug 5

    Other

    Other2011 International Joint Conference on Neural Network, IJCNN 2011
    CitySan Jose, CA
    Period11/7/3111/8/5

    Fingerprint

    Independent component analysis

    ASJC Scopus subject areas

    • Software
    • Artificial Intelligence

    Cite this

    Yokote, R., Nakamura, T., & Matsuyama, Y. (2011). Independent component analysis with graphical correlation: Applications to multi-vision coding. In Proceedings of the International Joint Conference on Neural Networks (pp. 701-708). [6033290] https://doi.org/10.1109/IJCNN.2011.6033290

    Independent component analysis with graphical correlation : Applications to multi-vision coding. / Yokote, Ryota; Nakamura, Toshikazu; Matsuyama, Yasuo.

    Proceedings of the International Joint Conference on Neural Networks. 2011. p. 701-708 6033290.

    Research output: Chapter in Book/Report/Conference proceedingConference contribution

    Yokote, R, Nakamura, T & Matsuyama, Y 2011, Independent component analysis with graphical correlation: Applications to multi-vision coding. in Proceedings of the International Joint Conference on Neural Networks., 6033290, pp. 701-708, 2011 International Joint Conference on Neural Network, IJCNN 2011, San Jose, CA, 11/7/31. https://doi.org/10.1109/IJCNN.2011.6033290
    Yokote R, Nakamura T, Matsuyama Y. Independent component analysis with graphical correlation: Applications to multi-vision coding. In Proceedings of the International Joint Conference on Neural Networks. 2011. p. 701-708. 6033290 https://doi.org/10.1109/IJCNN.2011.6033290
    Yokote, Ryota ; Nakamura, Toshikazu ; Matsuyama, Yasuo. / Independent component analysis with graphical correlation : Applications to multi-vision coding. Proceedings of the International Joint Conference on Neural Networks. 2011. pp. 701-708
    @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",

    }

    TY - GEN

    T1 - Independent component analysis with graphical correlation

    T2 - Applications to multi-vision coding

    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

    SN - 9781457710865

    SP - 701

    EP - 708

    BT - Proceedings of the International Joint Conference on Neural Networks

    ER -