Report

Representation of human faces

Author(s) / Creator(s)

Vetter, Thomas
Troje, Nikolaus F.

Abstract / Description

Several models for parameterized face representations have been proposed in the last years. A simple coding scheme treats the image of a face as a long vector with each entry coding for the intensity of one single pixel in the image. (e.g. Sirovich & Kirby 1987). Although simple and straitforward, such pixel-based representations have several disadvantages. We propose a representation for images of faces that separates texture and 2D shape by exploiting pixel-by-pixel correspondence between the images. The advantages of this representation compared to pixel-based representations are demonstrated by means of the quality of low-dimensional reconstructions derived from principal component analysis and by means of the performance that simple linear classifier can achieve for sex classification.

Keyword(s)

Gesicht Biometrie Optischer Fluss Gesicht Biometrie Optischer Fluss pixel-based representations human faces pixel-by-pixel correspondence

Persistent Identifier

Date of first publication

1996

Citation

  • Author(s) / Creator(s)
    Vetter, Thomas
  • Author(s) / Creator(s)
    Troje, Nikolaus F.
  • PsychArchives acquisition timestamp
    2022-11-17T11:03:33Z
  • Made available on
    2007-03-14
  • Made available on
    2015-12-01T10:32:05Z
  • Made available on
    2022-11-17T11:03:33Z
  • Date of first publication
    1996
  • Abstract / Description
    Several models for parameterized face representations have been proposed in the last years. A simple coding scheme treats the image of a face as a long vector with each entry coding for the intensity of one single pixel in the image. (e.g. Sirovich & Kirby 1987). Although simple and straitforward, such pixel-based representations have several disadvantages. We propose a representation for images of faces that separates texture and 2D shape by exploiting pixel-by-pixel correspondence between the images. The advantages of this representation compared to pixel-based representations are demonstrated by means of the quality of low-dimensional reconstructions derived from principal component analysis and by means of the performance that simple linear classifier can achieve for sex classification.
    en
  • Persistent Identifier
    https://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bsz:291-psydok-9315
  • Persistent Identifier
    https://hdl.handle.net/20.500.11780/1046
  • Persistent Identifier
    https://doi.org/10.23668/psycharchives.9000
  • Language of content
    eng
  • Keyword(s)
    Gesicht
    de
  • Keyword(s)
    Biometrie
    de
  • Keyword(s)
    Optischer Fluss
    de
  • Keyword(s)
    Gesicht
    de
  • Keyword(s)
    Biometrie
    de
  • Keyword(s)
    Optischer Fluss
    de
  • Keyword(s)
    pixel-based representations
    en
  • Keyword(s)
    human faces
    en
  • Keyword(s)
    pixel-by-pixel correspondence
    en
  • Dewey Decimal Classification number(s)
    150
  • Title
    Representation of human faces
    en
  • DRO type
    report
  • Visible tag(s)
    PsyDok