Regression based automatic face annotation for deformable model building

Akshay Asthana, Simon Lucey, Roland Goecke

    Research output: Contribution to journalArticle

    15 Citations (Scopus)
    2 Downloads (Pure)

    Abstract

    A major drawback of statistical models of non-rigid, deformable objects, such as the active appearance model (AAM), is the required pseudo-dense annotation of landmark points for every training image. We propose a regression-based approach for automatic annotation of face images at arbitrary pose and expression, and for deformable model building using only the annotated frontal images. We pose the problem of learning the pattern of manual annotation as a data-driven regression problem and explore several regression strategies to effectively predict the spatial arrangement of the landmark points for unseen face images, with arbitrary expression, at arbitrary poses. We show that the proposed fully sparse non-linear regression approach outperforms other regression strategies by effectively modelling the changes in the shape of the face under varying pose and is capable of capturing the subtleties of different facial expressions at the same time, thus, ensuring the high quality of the generated synthetic images. We show the generalisability of the proposed approach by automatically annotating the face images from four different databases and verifying the results by comparing them with a ground truth obtained from manual annotations
    Original languageEnglish
    Pages (from-to)2598-2613
    Number of pages16
    JournalPattern Recognition
    Volume44
    Issue number11
    DOIs
    Publication statusPublished - 2011

    Fingerprint

    Statistical Models

    Cite this

    Asthana, Akshay ; Lucey, Simon ; Goecke, Roland. / Regression based automatic face annotation for deformable model building. In: Pattern Recognition. 2011 ; Vol. 44, No. 11. pp. 2598-2613.
    @article{1504b18e13b84799ad5191d80c8459be,
    title = "Regression based automatic face annotation for deformable model building",
    abstract = "A major drawback of statistical models of non-rigid, deformable objects, such as the active appearance model (AAM), is the required pseudo-dense annotation of landmark points for every training image. We propose a regression-based approach for automatic annotation of face images at arbitrary pose and expression, and for deformable model building using only the annotated frontal images. We pose the problem of learning the pattern of manual annotation as a data-driven regression problem and explore several regression strategies to effectively predict the spatial arrangement of the landmark points for unseen face images, with arbitrary expression, at arbitrary poses. We show that the proposed fully sparse non-linear regression approach outperforms other regression strategies by effectively modelling the changes in the shape of the face under varying pose and is capable of capturing the subtleties of different facial expressions at the same time, thus, ensuring the high quality of the generated synthetic images. We show the generalisability of the proposed approach by automatically annotating the face images from four different databases and verifying the results by comparing them with a ground truth obtained from manual annotations",
    keywords = "Automatic face annotation, Deformable face model, Active appearance model",
    author = "Akshay Asthana and Simon Lucey and Roland Goecke",
    year = "2011",
    doi = "10.1016/J.PATCOG.2011.03.014",
    language = "English",
    volume = "44",
    pages = "2598--2613",
    journal = "Pattern Recognition",
    issn = "0031-3203",
    publisher = "Elsevier Limited",
    number = "11",

    }

    Regression based automatic face annotation for deformable model building. / Asthana, Akshay; Lucey, Simon; Goecke, Roland.

    In: Pattern Recognition, Vol. 44, No. 11, 2011, p. 2598-2613.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Regression based automatic face annotation for deformable model building

    AU - Asthana, Akshay

    AU - Lucey, Simon

    AU - Goecke, Roland

    PY - 2011

    Y1 - 2011

    N2 - A major drawback of statistical models of non-rigid, deformable objects, such as the active appearance model (AAM), is the required pseudo-dense annotation of landmark points for every training image. We propose a regression-based approach for automatic annotation of face images at arbitrary pose and expression, and for deformable model building using only the annotated frontal images. We pose the problem of learning the pattern of manual annotation as a data-driven regression problem and explore several regression strategies to effectively predict the spatial arrangement of the landmark points for unseen face images, with arbitrary expression, at arbitrary poses. We show that the proposed fully sparse non-linear regression approach outperforms other regression strategies by effectively modelling the changes in the shape of the face under varying pose and is capable of capturing the subtleties of different facial expressions at the same time, thus, ensuring the high quality of the generated synthetic images. We show the generalisability of the proposed approach by automatically annotating the face images from four different databases and verifying the results by comparing them with a ground truth obtained from manual annotations

    AB - A major drawback of statistical models of non-rigid, deformable objects, such as the active appearance model (AAM), is the required pseudo-dense annotation of landmark points for every training image. We propose a regression-based approach for automatic annotation of face images at arbitrary pose and expression, and for deformable model building using only the annotated frontal images. We pose the problem of learning the pattern of manual annotation as a data-driven regression problem and explore several regression strategies to effectively predict the spatial arrangement of the landmark points for unseen face images, with arbitrary expression, at arbitrary poses. We show that the proposed fully sparse non-linear regression approach outperforms other regression strategies by effectively modelling the changes in the shape of the face under varying pose and is capable of capturing the subtleties of different facial expressions at the same time, thus, ensuring the high quality of the generated synthetic images. We show the generalisability of the proposed approach by automatically annotating the face images from four different databases and verifying the results by comparing them with a ground truth obtained from manual annotations

    KW - Automatic face annotation

    KW - Deformable face model

    KW - Active appearance model

    U2 - 10.1016/J.PATCOG.2011.03.014

    DO - 10.1016/J.PATCOG.2011.03.014

    M3 - Article

    VL - 44

    SP - 2598

    EP - 2613

    JO - Pattern Recognition

    JF - Pattern Recognition

    SN - 0031-3203

    IS - 11

    ER -