All Points Considered: A Maximum Likelihood Method for Motion Recovery

  • Daniel Keren
  • Ilan Shimshoni
  • Liran Goshen
  • Michael Werman
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2616)


This paper addresses the problem of motion parameter recovery. A novel paradigm is offered to this problem, which computes a maximum likelihood (ML) estimate. The main novelty is that all domain-range point combinations are considered, as opposed to a single “optimal” combination. While this involves the optimization of nontrivial cost functions, the results are superior to those of the so-called algebraic and geometric methods, especially under the presence of strong noise, or when the measurement points approach a degenerate configuration.


Noise Variance Scatter Diagram Geometric Method Algebraic Method Geometric Estimate 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    A. Adam, E. Rivlin, and I. Shimshoni. Computingthe sensory uncertainty field of a vision-based localization sensor. IEEE Trans. on Robotics and Automation, 17(3):258–267, June 2001.CrossRefGoogle Scholar
  2. [2]
    D. A. Forsyth, S. Ioffe, and J. Haddon. Bayesian structure from motion. In ICCV99, pages 660–665, 1999.Google Scholar
  3. [3]
    R. Hartley and A. Zisserman. Multiple Views Geoemtry in Computer Vision. Cambridge University Press, 2000.Google Scholar
  4. [4]
    R. I. Hartley. In defense of the eight-point algorithm. IEEE Trans. Patt. Anal. Mach. Intell., 19(6):580–593, June 1997.CrossRefGoogle Scholar
  5. [5]
    K. Kanatani. Geometric computation for machine vision. In Oxford University Press, 1993.Google Scholar
  6. [6]
    K. Kanatani. Statistical-analysis of geometric computation. CVGIP, 59(3):286–306, May 1994.CrossRefGoogle Scholar
  7. [7]
    H. C. Longuet-Higgins. A computer algorithm for reconstructing a scene from two projections. Nature, 293:133–135, 1981.CrossRefGoogle Scholar
  8. [8]
    W. Press, B. Flannery, S. Teukolsky, and W. Vetterling. Numerical Recipes in C. Cambridge University Press, 1988.Google Scholar
  9. [9]
    G. Speyer and M. Werman. Parameter estimates for a pencil of lines: Bounds and estimators. In ECCV, 2002.Google Scholar
  10. [10]
    P. H. S. Torr and A. Zisserman. Concerning Bayesian motion segmentation, model averaging, matching and the trifocal tensor. In H. Burkharddt and B. Neumann, editors, ECCV98 Vol 1, pages 511–528. Springer, 1998.Google Scholar
  11. [11]
    M. Werman and D. Keren. A Bayesian method for fittingp arametric and nonparametric models to noisy data. IEEE Trans. Patt. Anal. Mach. Intell., 23(5):528–534, May 2001.CrossRefGoogle Scholar
  12. [12]
    Z. Zhang, R. Deriche, O. Faugeras, and Q. Luong. A robust technique for matching two uncalibrated images through the recovery of unknown epipolar geometry. Artificial Inteligence, 78(1–2):88–119, 1995.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Daniel Keren
    • 1
  • Ilan Shimshoni
    • 2
  • Liran Goshen
    • 2
  • Michael Werman
    • 3
  1. 1.Department of Computer ScienceUniversity of HaifaHaifaIsrael
  2. 2.Faculty of Industrial Engineering, TechnionTechnion CityIsrael
  3. 3.School of Computer Science and EngineeringThe Hebrew University of JerusalemJerusalemIsrael

Personalised recommendations