• DocumentCode
    2512458
  • Title

    Hyper Least Squares and Its Applications

  • Author

    Rangarajan, Prasanna ; Kanatani, Kenichi ; Niitsuma, Hirotaka ; Sugaya, Yasuyuki

  • fYear
    2010
  • fDate
    23-26 Aug. 2010
  • Firstpage
    5
  • Lastpage
    8
  • Abstract
    We present a new form of least squares (LS), called "hyper LS", for geometric problems that frequently appear in computer vision applications. Doing rigorous error analysis, we maximize the accuracy by introducing a normalization that eliminates statistical bias up to second order noise terms. Our method yields a solution comparable to maximum likelihood (ML) without iterations, even in large noise situations where ML computation fails.
  • Keywords
    least mean squares methods; statistical analysis; geometric problem; hyper least squares method; statistical bias; Accuracy; Covariance matrix; Equations; Error analysis; Maximum likelihood estimation; Noise; Transmission line matrix methods; ellipse fitting; fundamental matrix; geometric fitting; homography; least squares; maximum likelihood;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Pattern Recognition (ICPR), 2010 20th International Conference on
  • Conference_Location
    Istanbul
  • ISSN
    1051-4651
  • Print_ISBN
    978-1-4244-7542-1
  • Type

    conf

  • DOI
    10.1109/ICPR.2010.10
  • Filename
    5597662
    • Link To Document
    https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=2512458