3D Shape Representation by Fusing Local and Global Information

F.R.M. Al-Osaimi, Mohammed Bennamoun, Ajmal Mian

    Research output: Chapter in Book/Conference paperConference paper

    Abstract

    We present a unified feature representation of 2.5D pointclouds and apply it to face recognition. The representation integrates local and global geometrical cues in a single compact representation using tensor fields. The global cues provide geometrical coherence for the local cues resulting in better descriptiveness of the unified representation. Multiple rank-0 tensor fields are computed at every point from its local neighborhood and from the global structure of the 2.5D pointcloud. The pointcloud is then represented by multiple rank-0 tensor fields which are invariant to rigid transformations. Each local tensor field is integrated with every global field in a 2D histogram which is indexed by a local field in one dimension and a global field in the other dimension. Finally, PCA coefficients of the 2D histograms are concatenated into a single feature vector. The representation was tested on FRGC V2.0 dataset and achieved 93.78% identification rate and 95.37% verification rate at 0.1% FAR.
    Original languageEnglish
    Title of host publicationProceedings of 9th International Symposium on Signal Processing and Its Applications, ISSPA 2007
    Place of PublicationLos Alamitos, California, USA
    PublisherIEEE, Institute of Electrical and Electronics Engineers
    Pages1-4
    DOIs
    Publication statusPublished - 2007
    Event3D Shape Representation by Fusing Local and Global Information - Sharjah, United Arab Emirates
    Duration: 1 Jan 2007 → …

    Conference

    Conference3D Shape Representation by Fusing Local and Global Information
    Period1/01/07 → …

    Fingerprint Dive into the research topics of '3D Shape Representation by Fusing Local and Global Information'. Together they form a unique fingerprint.

    Cite this