• DocumentCode
    702352
  • Title

    Influence of pose angle on face recognition from very low resolution images

  • Author

    Marciniak, Tomasz ; Weychan, Radoslaw ; Chmielewska, Agata ; Dabrowski, Adam

  • Author_Institution
    Div. of Signal Process. & Electron. Syst., Poznan Univ. of Technol., Poznań, Poland
  • fYear
    2012
  • fDate
    27-29 Sept. 2012
  • Firstpage
    177
  • Lastpage
    181
  • Abstract
    This paper presents analysis of impact of the image resolution on the efficacy of the automatic face recognition. During experimental studies three databases were tested, where the head in the photos is set at different angles. The effectiveness of face location detection was examined with the use of the skin color and geometric models. Next, we tested the influence of the head position and the image resolution on the people identification effectiveness. Effect of lowering the image resolution is presented in a FAR / FRR graph form.
  • Keywords
    computational geometry; face recognition; image colour analysis; image resolution; pose estimation; FAR-FRR graph form; automatic face recognition; face location detection; geometric model; head position; people identification; pose angle; skin color model; very-low resolution images; Databases; Decision support systems; Face; Face recognition; Image resolution; Signal resolution; face databases; face detection; face recognition; low resolution;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    New Trends in Audio & Video and Signal Processing: Algorithms, Architectures, Arrangements, and Applications (NTAV/SPA), 2012 Joint Conference
  • Conference_Location
    Lodz
  • Print_ISBN
    978-8-3728-3502-4
  • Type

    conf

  • Filename
    7085532