• DocumentCode
    2867457
  • Title

    The Judgment Method for the Unique Solution of Real-time Pose Estimation from Particular Line Correspondences

  • Author

    QIN, Lijuan ; Zhu, Feng

  • Author_Institution
    Shenyang Inst. of Autom., Chinese Acad. of Sci., Shenyang
  • fYear
    2006
  • fDate
    25-28 June 2006
  • Firstpage
    1216
  • Lastpage
    1220
  • Abstract
    We usually can determine object pose from three lines in general position. For the particular line configuration of three non-coplanar lines which intersect at two points, it has some particular characteristics three lines in general position don´t have, thus its mathematical expression is easy and it is easy to solve if we use this particular line configuration to locate object. This method enriches the pose estimation methods from three line correspondences in theory, at the same time, it provides the base for the practical applications. The article introduces the pose estimation method from this particular line configuration. The multi-solution phenomenon is the dominating problem that puzzles the computer vision for a long time and it limits the application of pose estimation algorithm. To solve the multi-solution problem, we present a novel method to find the unique solution by position information of space lines. Especially, we provide a detailed account of the realizable aspects of the judgment method. At last, the simulation results which demonstrate our algorithm method works fast and robust are presented
  • Keywords
    computer vision; motion estimation; object detection; computer vision; multisolution phenomenon; noncoplanar lines; object pose; particular line configuration; real-time pose estimation; Application software; Automation; Computer vision; Mechatronics; Navigation; Nonlinear equations; Object recognition; Polynomials; Robots; Robustness; Line features; Multi-solution phenomenon; Pose estimation; Unique solution;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Mechatronics and Automation, Proceedings of the 2006 IEEE International Conference on
  • Conference_Location
    Luoyang, Henan
  • Print_ISBN
    1-4244-0465-7
  • Electronic_ISBN
    1-4244-0466-5
  • Type

    conf

  • DOI
    10.1109/ICMA.2006.257799
  • Filename
    4026259