Occlusion-robust 3D face recognition using restoration and local classifiers

Nese Alyuz, B. Gökberk, Lieuwe Jan Spreeuwers, Raymond N.J. Veldhuis, Lale Akarun

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

    Abstract

    Occlusions complicate the process of identifying individuals using their 3D facial scans. We propose a 3D face recognition system that automatically removes occlusion artifacts and identifies the facial image using regional classifiers. Automatic localization of occluded areas is handled by using a generic face model. Restoration of missing information after occlusion removal is performed by the application of an improved version of Gappy Principal Component Analysis (GPCA), which we call partial Gappy PCA (pGPCA). After the removal of noisy data introduced by realistic occlusions, occlusion-free faces are represented by local regions. Local classifiers operating on these local regions are then fused to achieve occlusion-robust identification performance. Our experimental results obtained on realistically occluded facial images from the Bosphorus 3D face database illustrate that our occlusion compensation scheme drastically improves the recognition accuracy from 78.05% to 94.20%.
    Original languageUndefined
    Title of host publicationProceedings of the 2011 IEEE 19th Conference on Signal Processing and Communications Applications (SIU 2011)
    Place of PublicationUSA
    PublisherIEEE Signal Processing Society
    Pages750-753
    Number of pages4
    ISBN (Print)978-1-4577-0462-8
    DOIs
    Publication statusPublished - 20 Apr 2011

    Publication series

    Name
    PublisherIEEE Signal Processing Society

    Keywords

    • IR-78805
    • METIS-281627
    • occlusion-robust 3D face recognition
    • occlusion artifact
    • Gappy principal component analysis
    • GPCA
    • restoration classifier
    • image identification
    • automatic localization
    • missing information Restoration
    • local classifier
    • partial Gappy PCA
    • pGPCA
    • generic face model
    • SCS-Safety
    • EWI-20961

    Cite this