8th International Conference on Bio-inspired Information and Communications Technologies (formerly BIONETICS)

Research Article

SVD-based Feature Extraction from Time-series Motion Data and Its Application to Gesture Recognition

  • @INPROCEEDINGS{10.4108/icst.bict.2014.257811,
        author={Isao Hayashi and Yinlai Jiang and Shuoyu Wang},
        title={SVD-based Feature Extraction from Time-series Motion Data and Its Application to Gesture Recognition},
        proceedings={8th International Conference on Bio-inspired Information and Communications Technologies (formerly BIONETICS)},
        publisher={ICST},
        proceedings_a={BICT},
        year={2015},
        month={2},
        keywords={feature extraction svd gesture recognition},
        doi={10.4108/icst.bict.2014.257811}
    }
    
  • Isao Hayashi
    Yinlai Jiang
    Shuoyu Wang
    Year: 2015
    SVD-based Feature Extraction from Time-series Motion Data and Its Application to Gesture Recognition
    BICT
    ACM
    DOI: 10.4108/icst.bict.2014.257811
Isao Hayashi1, Yinlai Jiang,*, Shuoyu Wang2
  • 1: Kansai University
  • 2: Kochi University of Technology
*Contact email: jiang@hi.mce.uec.ac.jp

Abstract

Singular value decomposition is used to extract features from time-series motion data. A matrix consisting of the time-series data is decomposed into left singular vectors which represent the patterns of the motion and singular values as a scalar, by which each corresponding left singular vector affects the matrix. Gesture recognition using the extracted features suggest the effectiveness of the method.