Comparison of Silhouette Shape Descriptors for Example-based Human Pose Recovery

Ronald Walter Poppe, Mannes Poel

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    39 Citations (Scopus)

    Abstract

    Automatically recovering human poses from visual input is useful but challenging due to variations in image space and the high dimensionality of the pose space. In this paper, we assume that a human silhouette can be extracted from monocular visual input. We compare three shape descriptors that are used in the encoding of silhouettes: Fourier descriptors, shape contexts and Hu moments. An examplebased approach is taken to recover upper body poses from these descriptors. We perform experiments with deformed silhouettes to test each descriptor’s robustness against variations in body dimensions, viewpoint and noise. It is shown that Fourier descriptors and shape context histograms outperform Hu moments for all deformations.
    Original languageUndefined
    Title of host publicationProceedings of the IEEE Conference on Automatic Face and Gesture Recognition 2006 (FG 2006)
    Place of PublicationLos Alamitos
    PublisherIEEE Computer Society
    Pages541-546
    Number of pages6
    ISBN (Print)0-7695-2503-2
    DOIs
    Publication statusPublished - 10 Apr 2006
    Event7th International Conference on Automatic Face and Gesture Recognition, FG 2006 - Southhamton, United Kingdom
    Duration: 10 Apr 200612 Apr 2006
    Conference number: 7
    http://www.fg2006.ecs.soton.ac.uk/

    Publication series

    Name
    PublisherIEEE Computer Society Press
    Number2

    Conference

    Conference7th International Conference on Automatic Face and Gesture Recognition, FG 2006
    Abbreviated titleFG
    CountryUnited Kingdom
    CitySouthhamton
    Period10/04/0612/04/06
    Internet address

    Keywords

    • METIS-238153
    • EC Grant Agreement nr.: FP6/506811
    • EWI-6866
    • IR-63418

    Cite this

    Poppe, R. W., & Poel, M. (2006). Comparison of Silhouette Shape Descriptors for Example-based Human Pose Recovery. In Proceedings of the IEEE Conference on Automatic Face and Gesture Recognition 2006 (FG 2006) (pp. 541-546). [10.1109/FGR.2006.32] Los Alamitos: IEEE Computer Society. https://doi.org/10.1109/FGR.2006.32
    Poppe, Ronald Walter ; Poel, Mannes. / Comparison of Silhouette Shape Descriptors for Example-based Human Pose Recovery. Proceedings of the IEEE Conference on Automatic Face and Gesture Recognition 2006 (FG 2006). Los Alamitos : IEEE Computer Society, 2006. pp. 541-546
    @inproceedings{4102b9fab4464f129f584ce8381a1402,
    title = "Comparison of Silhouette Shape Descriptors for Example-based Human Pose Recovery",
    abstract = "Automatically recovering human poses from visual input is useful but challenging due to variations in image space and the high dimensionality of the pose space. In this paper, we assume that a human silhouette can be extracted from monocular visual input. We compare three shape descriptors that are used in the encoding of silhouettes: Fourier descriptors, shape contexts and Hu moments. An examplebased approach is taken to recover upper body poses from these descriptors. We perform experiments with deformed silhouettes to test each descriptor’s robustness against variations in body dimensions, viewpoint and noise. It is shown that Fourier descriptors and shape context histograms outperform Hu moments for all deformations.",
    keywords = "METIS-238153, EC Grant Agreement nr.: FP6/506811, EWI-6866, IR-63418",
    author = "Poppe, {Ronald Walter} and Mannes Poel",
    note = "10.1109/FGR.2006.32",
    year = "2006",
    month = "4",
    day = "10",
    doi = "10.1109/FGR.2006.32",
    language = "Undefined",
    isbn = "0-7695-2503-2",
    publisher = "IEEE Computer Society",
    number = "2",
    pages = "541--546",
    booktitle = "Proceedings of the IEEE Conference on Automatic Face and Gesture Recognition 2006 (FG 2006)",
    address = "United States",

    }

    Poppe, RW & Poel, M 2006, Comparison of Silhouette Shape Descriptors for Example-based Human Pose Recovery. in Proceedings of the IEEE Conference on Automatic Face and Gesture Recognition 2006 (FG 2006)., 10.1109/FGR.2006.32, IEEE Computer Society, Los Alamitos, pp. 541-546, 7th International Conference on Automatic Face and Gesture Recognition, FG 2006, Southhamton, United Kingdom, 10/04/06. https://doi.org/10.1109/FGR.2006.32

    Comparison of Silhouette Shape Descriptors for Example-based Human Pose Recovery. / Poppe, Ronald Walter; Poel, Mannes.

    Proceedings of the IEEE Conference on Automatic Face and Gesture Recognition 2006 (FG 2006). Los Alamitos : IEEE Computer Society, 2006. p. 541-546 10.1109/FGR.2006.32.

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    TY - GEN

    T1 - Comparison of Silhouette Shape Descriptors for Example-based Human Pose Recovery

    AU - Poppe, Ronald Walter

    AU - Poel, Mannes

    N1 - 10.1109/FGR.2006.32

    PY - 2006/4/10

    Y1 - 2006/4/10

    N2 - Automatically recovering human poses from visual input is useful but challenging due to variations in image space and the high dimensionality of the pose space. In this paper, we assume that a human silhouette can be extracted from monocular visual input. We compare three shape descriptors that are used in the encoding of silhouettes: Fourier descriptors, shape contexts and Hu moments. An examplebased approach is taken to recover upper body poses from these descriptors. We perform experiments with deformed silhouettes to test each descriptor’s robustness against variations in body dimensions, viewpoint and noise. It is shown that Fourier descriptors and shape context histograms outperform Hu moments for all deformations.

    AB - Automatically recovering human poses from visual input is useful but challenging due to variations in image space and the high dimensionality of the pose space. In this paper, we assume that a human silhouette can be extracted from monocular visual input. We compare three shape descriptors that are used in the encoding of silhouettes: Fourier descriptors, shape contexts and Hu moments. An examplebased approach is taken to recover upper body poses from these descriptors. We perform experiments with deformed silhouettes to test each descriptor’s robustness against variations in body dimensions, viewpoint and noise. It is shown that Fourier descriptors and shape context histograms outperform Hu moments for all deformations.

    KW - METIS-238153

    KW - EC Grant Agreement nr.: FP6/506811

    KW - EWI-6866

    KW - IR-63418

    U2 - 10.1109/FGR.2006.32

    DO - 10.1109/FGR.2006.32

    M3 - Conference contribution

    SN - 0-7695-2503-2

    SP - 541

    EP - 546

    BT - Proceedings of the IEEE Conference on Automatic Face and Gesture Recognition 2006 (FG 2006)

    PB - IEEE Computer Society

    CY - Los Alamitos

    ER -

    Poppe RW, Poel M. Comparison of Silhouette Shape Descriptors for Example-based Human Pose Recovery. In Proceedings of the IEEE Conference on Automatic Face and Gesture Recognition 2006 (FG 2006). Los Alamitos: IEEE Computer Society. 2006. p. 541-546. 10.1109/FGR.2006.32 https://doi.org/10.1109/FGR.2006.32