• DocumentCode
    2033922
  • Title

    Object based 3-D motion and structure estimation

  • Author

    Alatan, A. Aydzn ; Onural, Levent

  • Author_Institution
    Dept. of Electr. & Electron. Eng., Bilkent Univ., Ankara, Turkey
  • Volume
    1
  • fYear
    1995
  • fDate
    23-26 Oct 1995
  • Firstpage
    390
  • Abstract
    Motion analysis is the most crucial part of object-based coding. Motion in a 3-D environment can be analyzed better by using a 3-D motion model compared to its 2-D counterpart and hence may improve coding efficiency. Gibbs formulated joint segmentation and estimation of 2-D motion not only improves the performance, but also generates robust point correspondences which are necessary for linear 3-D motion estimation algorithms. Estimated 3-D motion parameters are used to find the structure of the previously segmented objects by minimizing another Gibbs energy. Such an approach achieves error immunity compared to linear algorithms. Experimental results are promising and hence the proposed motion and structure analysis method is a candidate to be used in object-based (or even knowledge-based) video coding schemes
  • Keywords
    Bayes methods; image segmentation; motion estimation; parameter estimation; video coding; 2D motion segmentation; 3D motion estimation; 3D motion parameters; 3D structure estimation; Bayesian approach; Gibbs energy; coding efficiency; error immunity; experimental results; knowledge based video coding; linear 3D motion estimation algorithms; motion analysis; object based coding; object based video coding; parameter estimation; performance; robust point correspondences; structure analysis method; Immune system; Layout; Motion analysis; Motion compensation; Motion estimation; Object segmentation; Parameter estimation; Robustness; Video coding; Video compression;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Image Processing, 1995. Proceedings., International Conference on
  • Conference_Location
    Washington, DC
  • Print_ISBN
    0-8186-7310-9
  • Type

    conf

  • DOI
    10.1109/ICIP.1995.529728
  • Filename
    529728