• DocumentCode
    2025353
  • Title

    Applying Two-Pixel Features to Face Detection

  • Author

    Nissenboim, Ido ; Keren, Daniel ; Werman, Michael

  • Author_Institution
    Comput. Sci. Dept., Univ. of Haifa, Haifa, Israel
  • fYear
    2008
  • fDate
    Nov. 30 2008-Dec. 3 2008
  • Firstpage
    65
  • Lastpage
    69
  • Abstract
    We propose a "quick rejection" paradigm which is applied to face detection. The features we use are arguably the simplest possible: a threshold on the difference between the grey levels of two pixels. No negative examples are used for training; instead, we use a simple statistical model of natural images. The resulting features are easy to find, extremely fast to apply, and achieve a good detection rate.
  • Keywords
    face recognition; feature extraction; statistical analysis; face detection; grey levels; pixel features; statistical model; Boosting; Computer science; Computer vision; Convolution; Detectors; Face detection; Humans; Internet; Pixel; Testing; fast image detection;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Signal Image Technology and Internet Based Systems, 2008. SITIS '08. IEEE International Conference on
  • Conference_Location
    Bali
  • Print_ISBN
    978-0-7695-3493-0
  • Type

    conf

  • DOI
    10.1109/SITIS.2008.28
  • Filename
    4725788