Biometric identification using motion history images of a speaker's lip movements

Alfredo Grunwald de la Cuesta, Jianguo Zhang, Paul Miller

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

    13 Citations (Scopus)

    Abstract

    This paper describes a new simple, but effective, approach to speaker verification using video sequences of lip movements. We use motion history images (MHI) to provide a biometric template of a spoken word for each speaker. Class-dependent correlation filters are then created by a weighted optimization of training MHI samples. Feature extraction is performed by correlating a test MHI against each correlation filter. A Bayesian classifier is deployed for classification. We carry out an extensive performance evaluation of our approach with respect to the number of training samples and different words. Results clearly show the efficacy of our method.
    Original languageEnglish
    Title of host publicationProceedings 2008 International Machine Vision and Image Processing Conference
    Subtitle of host publicationIMVIP 2008
    EditorsBryan Scotney, Philip Morrow
    Place of PublicationLos Alamitos, Calif.
    PublisherIEEE
    Pages83-88
    Number of pages6
    ISBN (Print)9780769533322
    DOIs
    Publication statusPublished - 2008
    EventIMVIP 2008 International Machine Vision and Image Processing Conference - University of Ulster, Portrush, Coleraine, Ireland
    Duration: 3 Sep 20085 Sep 2008
    http://www.compeng.ulster.ac.uk/events/imvip2008/

    Conference

    ConferenceIMVIP 2008 International Machine Vision and Image Processing Conference
    Abbreviated titleIMVIP 2008
    CountryIreland
    CityPortrush, Coleraine
    Period3/09/085/09/08
    Internet address

    Fingerprint

    Biometrics
    Feature extraction
    Classifiers

    Cite this

    de la Cuesta, A. G., Zhang, J., & Miller, P. (2008). Biometric identification using motion history images of a speaker's lip movements. In B. Scotney, & P. Morrow (Eds.), Proceedings 2008 International Machine Vision and Image Processing Conference: IMVIP 2008 (pp. 83-88). Los Alamitos, Calif.: IEEE. https://doi.org/10.1109/IMVIP.2008.13
    de la Cuesta, Alfredo Grunwald ; Zhang, Jianguo ; Miller, Paul. / Biometric identification using motion history images of a speaker's lip movements. Proceedings 2008 International Machine Vision and Image Processing Conference: IMVIP 2008. editor / Bryan Scotney ; Philip Morrow. Los Alamitos, Calif. : IEEE, 2008. pp. 83-88
    @inproceedings{3999eb9509224f54b5faf67f98da3c7c,
    title = "Biometric identification using motion history images of a speaker's lip movements",
    abstract = "This paper describes a new simple, but effective, approach to speaker verification using video sequences of lip movements. We use motion history images (MHI) to provide a biometric template of a spoken word for each speaker. Class-dependent correlation filters are then created by a weighted optimization of training MHI samples. Feature extraction is performed by correlating a test MHI against each correlation filter. A Bayesian classifier is deployed for classification. We carry out an extensive performance evaluation of our approach with respect to the number of training samples and different words. Results clearly show the efficacy of our method.",
    author = "{de la Cuesta}, {Alfredo Grunwald} and Jianguo Zhang and Paul Miller",
    note = "Best Paper Award",
    year = "2008",
    doi = "10.1109/IMVIP.2008.13",
    language = "English",
    isbn = "9780769533322",
    pages = "83--88",
    editor = "Bryan Scotney and Morrow, {Philip }",
    booktitle = "Proceedings 2008 International Machine Vision and Image Processing Conference",
    publisher = "IEEE",

    }

    de la Cuesta, AG, Zhang, J & Miller, P 2008, Biometric identification using motion history images of a speaker's lip movements. in B Scotney & P Morrow (eds), Proceedings 2008 International Machine Vision and Image Processing Conference: IMVIP 2008. IEEE, Los Alamitos, Calif., pp. 83-88, IMVIP 2008 International Machine Vision and Image Processing Conference, Portrush, Coleraine, Ireland, 3/09/08. https://doi.org/10.1109/IMVIP.2008.13

    Biometric identification using motion history images of a speaker's lip movements. / de la Cuesta, Alfredo Grunwald; Zhang, Jianguo; Miller, Paul.

    Proceedings 2008 International Machine Vision and Image Processing Conference: IMVIP 2008. ed. / Bryan Scotney; Philip Morrow. Los Alamitos, Calif. : IEEE, 2008. p. 83-88.

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

    TY - GEN

    T1 - Biometric identification using motion history images of a speaker's lip movements

    AU - de la Cuesta, Alfredo Grunwald

    AU - Zhang, Jianguo

    AU - Miller, Paul

    N1 - Best Paper Award

    PY - 2008

    Y1 - 2008

    N2 - This paper describes a new simple, but effective, approach to speaker verification using video sequences of lip movements. We use motion history images (MHI) to provide a biometric template of a spoken word for each speaker. Class-dependent correlation filters are then created by a weighted optimization of training MHI samples. Feature extraction is performed by correlating a test MHI against each correlation filter. A Bayesian classifier is deployed for classification. We carry out an extensive performance evaluation of our approach with respect to the number of training samples and different words. Results clearly show the efficacy of our method.

    AB - This paper describes a new simple, but effective, approach to speaker verification using video sequences of lip movements. We use motion history images (MHI) to provide a biometric template of a spoken word for each speaker. Class-dependent correlation filters are then created by a weighted optimization of training MHI samples. Feature extraction is performed by correlating a test MHI against each correlation filter. A Bayesian classifier is deployed for classification. We carry out an extensive performance evaluation of our approach with respect to the number of training samples and different words. Results clearly show the efficacy of our method.

    U2 - 10.1109/IMVIP.2008.13

    DO - 10.1109/IMVIP.2008.13

    M3 - Conference contribution

    SN - 9780769533322

    SP - 83

    EP - 88

    BT - Proceedings 2008 International Machine Vision and Image Processing Conference

    A2 - Scotney, Bryan

    A2 - Morrow, Philip

    PB - IEEE

    CY - Los Alamitos, Calif.

    ER -

    de la Cuesta AG, Zhang J, Miller P. Biometric identification using motion history images of a speaker's lip movements. In Scotney B, Morrow P, editors, Proceedings 2008 International Machine Vision and Image Processing Conference: IMVIP 2008. Los Alamitos, Calif.: IEEE. 2008. p. 83-88 https://doi.org/10.1109/IMVIP.2008.13