Body-part templates for recovery of 2D human poses under occlusion

Ronald Walter Poppe, Mannes Poel

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

    4 Citations (Scopus)
    197 Downloads (Pure)


    Detection of humans and estimation of their 2D poses from a single image are challenging tasks. This is especially true when part of the observation is occluded. However, given a limited class of movements, poses can be recovered given the visible body-parts. To this end, we propose a novel template representation where the body is divided into five body-parts. Given a match, we not only estimate the joints in the body-part, but all joints in the body. Quantitative evaluation on a HumanEva walking sequence shows mean 2D errors of approximately 27.5 pixels. For simulated occlusion of the head and arms, similar results are obtained while occlusion of the legs increases this error by 6 pixels.
    Original languageUndefined
    Title of host publicationProceedings of the International Conference on Articulated Motion and Deformable Objects (AMDO)
    EditorsF.J. Perales, R.B. Fisher
    Place of PublicationBerlin
    Number of pages10
    ISBN (Print)978-3-540-70516-1
    Publication statusPublished - 27 Jun 2008
    EventInternational Conference on Articulated Motion and Deformable Objects, AMDO - Port d’Andratx, Spain
    Duration: 9 Jul 200811 Jul 2008

    Publication series

    NameLecture Notes in Computer Science
    PublisherSpringer Verlag
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349


    ConferenceInternational Conference on Articulated Motion and Deformable Objects, AMDO
    OtherJuly 9-11, 2008


    • EWI-12886
    • EC Grant Agreement nr.: FP6/033812
    • METIS-251011
    • IR-64813

    Cite this