Sparse Similarity-Based Fisherfaces

Jens Fagertun, David Delgado Gomez, Mads Fogtmann Hansen, Rasmus Reinhold Paulsen

    Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review


    In this work, the effect of introducing Sparse Principal Component Analysis within the Similarity-based Fisherfaces algorithm is examined. The technique aims at mimicking the human ability to discriminate faces by projecting the faces in a highly discriminative and easy interpretative way. Pixel intensities are used by Sparse Principal Component Analysis and Fisher Linear Discriminant Analysis to assign a one dimensional subspace projection to each person belonging to a reference data set. Experimental results performed in the AR dataset show that Similarity-based Fisherfaces in a sparse version can obtain the same recognition results as the technique in a dense version using only a fraction of the input data. Furthermore, the presented results suggest that using SPCA in the technique offers robustness to occlusions.
    Original languageEnglish
    Title of host publicationImage Analysis : 17th Scandinavian Conference, SCIA 2011 - Ystad, Sweden, May 2011 - Proceedings
    Publication date2011
    ISBN (Print)978-3-642-21226-0
    Publication statusPublished - 2011
    Event17th Scandinavian Conference on Image Analysis - Ystad, Sweden
    Duration: 23 May 201127 May 2011
    Conference number: 17


    Conference17th Scandinavian Conference on Image Analysis
    Internet address
    SeriesLecture Notes in Computer Science


    • Face recognition
    • Fisher Linear Discriminant Analysis
    • Biometrics
    • Sparse Principal Component Analysis
    • Multi- Subspace Method

    Fingerprint Dive into the research topics of 'Sparse Similarity-Based Fisherfaces'. Together they form a unique fingerprint.

    Cite this