Biological Cybernetics

, Volume 67, Issue 3, pp 259–268 | Cite as

Tracking facilitates 3-D motion estimation

  • Cornelia Fermüller
  • Yiannis Aloimonos
Article

Abstract

The recently emerging paradigm of Active Vision advocates studying visual problems in form of modules that are directly related to a visual task for observers that are active. Along these lines, we are arguing that in many cases when an object is moving in an unrestricted manner (translation and rotation) in the 3D world, we are just interested in the motion's translational components. For a monocular observer, using only the normal flow — the spatio-temporal derivatives of the image intensity function — we solve the problem of computing the direction of translation and the time to collision. We do not use optical flow since its computation is an ill-posed problem, and in the general case it is not the same as the motion field — the projection of 3D motion on the image plane. The basic idea of our motion parameter estimation strategy lies in the employment of fixation and tracking. Fixation simplifies much of the computation by placing the object at the center of the visual field, and the main advantage of tracking is the accumulation of information over time. We show how tracking is accomplished using normal flow measurements and use it for two different tasks in the solution process. First it serves as a tool to compensate for the lack of existence of an optical flow field and thus to estimate the translation parallel to the image plane; and second it gathers information about the motion component perpendicular to the image plane.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Adiv G (1985) Determining 3D-motion and structure from optical flow generated by serveral moving objects. IEEE Trans Pattern Anal Machine Intell PAMI-7:384–401Google Scholar
  2. Aloimonos JY (1990) Purposive and qualitative active vision. Proc. DARPA Image Understanding Workshop, pp 816–828Google Scholar
  3. Aloimonos J, Brown CM (1984) Direct processing of curvilinear sensor motion from a sequence of perspective images. In Proc Workshop on Computer Vision: Representation and Control, pp 72–77Google Scholar
  4. Aloimonos J, Brown CM (1989) On the kinetic depth effect. Biol Cybern 60:445–455Google Scholar
  5. Aloimonos J, Weiss I, Bandopadhay A (1988) Active vision. Int J Comput Vision 2:333–356Google Scholar
  6. Bajcsy R (1985) Active perceptive vs. passive perception. In Proc. IEEE Workshop on Computer Vision, pp 55–59Google Scholar
  7. Bandopadhay A, Ballard DH (1991) Egomotion perception using visual tracking. Comput Intell 7:39–47Google Scholar
  8. Fermüller C, Aloimonos Y (1991) Estimating 3-D Motion from Image Gradients. Technical Report CAR-TR-564, Center for Automation Research, University of MarylandGoogle Scholar
  9. Horn B, Schunck B (1981) Determining optical flow. Artif Intell 17:185–203Google Scholar
  10. Horn BKP, Weldon J (1987) Computational-efficient methods for recovering translational motion. In Proc International Conference on Computer Vision, pp 2–11Google Scholar
  11. Koenderink J (1986) Optic flow. Vision Res 26:161–180Google Scholar
  12. Koenderink J, van Doorn A (1975) Invariant properties of the motion parallax field due to the movement of rigid bodies relative to an observer. Opt Acta 22:773–791Google Scholar
  13. Longuet-Higgins HC, Prazdny K (1980) The interpretation of a moving retinal image. Proc R Soc London B 208:385–397Google Scholar
  14. Marr D (1982) Vision. Freeman, San FranciscoGoogle Scholar
  15. Negahdaripour S (1986) Direct passive navigation. PhD thesis, Department of Mechanical Engineering, MIT, Cambridge, MassGoogle Scholar
  16. Sharma R, Aloimonos J (1991) Robust detection of independent motion: an active and purposive solution. Technical Report CAR-TR-534, Center for Automation Research, University of MarylandGoogle Scholar
  17. Spetsakis ME, Aloimonos J (1990) Structure from motion using line correspondences. Int J Comput Vision 4:171–183Google Scholar
  18. Spetsakis ME, Aloimonos J (1988) Optimal computing of structure from motion using point correspondence. In Proc International Conference on Computer Vision, pp 449–453Google Scholar
  19. Taalebi-Nezhaad MA (1990) Direct recovery of motion and shape in the general case by fixation. In Proc DARPA Image Understanding Workshop, pp 284–291Google Scholar
  20. Tsai RY, Huang TS (1984) Uniqueness and estimation of three-dimensional motion parameters of rigid objects with curved surfaces. IEEE Trans Pattern Anal Machine Intell PAMI-6:13–27Google Scholar
  21. Ullman S (1979) The interpretation of visual motion. MIT Press, Cambridge, MassGoogle Scholar
  22. Verri A, Poggio T (1987) Against quantitative optic flow. Proc IEEE International Conference on Computer VisionGoogle Scholar
  23. Waxman AM Kamgar-Parsi B, Subbarao (1987) Closed-form solution to image flow equations for 3D structure and motion. Int J Comput Vision 1: 239–258Google Scholar

Copyright information

© Springer-Verlag 1992

Authors and Affiliations

  • Cornelia Fermüller
    • 1
    • 2
  • Yiannis Aloimonos
    • 1
  1. 1.Computer Vision Laboratory, Center for Automation ResearchUniversity of MarylandCollege ParkUSA
  2. 2.Department for Pattern Recognition and Image ProcessingInstitute for Automation, Technical University ViennaViennaAustria

Personalised recommendations