Decoupling the 3D motion space by fixation

  • Konstantinos Daniilidis
  • Inigo Thomas
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1064)


Fixation is defined as the ability of an active visual system to keep the projection of an environmental point stationary in the image. We show in this paper that fixation enables the decoupling of the 3D-motion parameters by projecting appropriately the spherical motion field in two latitudinal directions with respect to two different poles of the image sphere. Both computational steps are based on one-dimensional searches along meridians of the image sphere. We do not use the efference copy of the fixational rotation of the camera. Performance of the algorithm is tested on real world sequences with fixation accomplished either off-line or during the recording using an active camera.


Optical Flow Motion Field Translation Direction Efference Copy Corrective Saccade 
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.


  1. 1.
    R.H.S. Carpenter. Movements of the Eyes. Pion Press, London, 1988.Google Scholar
  2. 2.
    K.J. Bradshaw, P.F. McLauchlan, I.D. Reid, and D.W. Murray. Saccade and pursuit on an active head-eye platform. Image and Vision Computing, 12:155–163, 1994.Google Scholar
  3. 3.
    A. Bandopadhay and D.H. Ballard. Egomotion perception using visual tracking. Computational Intelligence, 7:39–47, 1990.Google Scholar
  4. 4.
    Y. Aloimonos, I. Weiss, and A. Bandyopadhyay. Active Vision. In Proc. Int. Conf. on Computer Vision, pages 35–54, London, UK, June 8–11, 1987.Google Scholar
  5. 5.
    C. Fermüller and Y. Aloimonos. The role of fixation in visual motion anaylsis. International Journal of Computer Vision, 11:165–186, 1993.Google Scholar
  6. 6.
    M.A. Taalebinezhaad. Direct recovery of motion and shape in the general case by fixation. IEEE Trans. Pattern Analysis and Machine Intelligence, 14:847–853, 1992.Google Scholar
  7. 7.
    D. Raviv and M. Herman. A unified approach to camera fixation and vision-based road following. IEEE Trans. Systems, Man, and Cybernetics, 24:1125–1141, 1994.Google Scholar
  8. 8.
    I. Thomas, E. Simoncelli, and R. Bajcsy. Spherical retinal flow for a fixating observer. In Proc. IEEE Workshop on Visual Behaviors, pages 37–41, 1994.Google Scholar
  9. 9.
    M. Tistarelli and G. Sandini. Dynamic aspects in active vision. CVGIP: Image Understanding, 56:108–129, 1992.Google Scholar
  10. 10.
    K. Daniilidis. Computation of 3D-motion parameters using the log-polar transform. In V. Hlavac et al. (Ed.), Proc. Int. Conf. Computer Analysis of Images and Patterns CAIP, Prag, pages 82–89, 1995.Google Scholar
  11. 11.
    M.J. Barth and S. Tsuji. Egomotion determination through an intelligent gaze control strategy. IEEE Trans. Systems, Man, and Cybernetics, 23:1424–1432, 1993.Google Scholar
  12. 12.
    V. Sundareswaran, P. Bouthemy, and F. Chaumette. Active camera self-orientation using dynamic image parameters. In Proc. Third European Conference on Computer Vision, pages 111–115. Stockholm, Sweden, May 2–6, J.O. Eklundh (Ed.), Springer LNCS 800, 1994.Google Scholar
  13. 13.
    S. Maybank. Theory of Reconstruction from Image Motion. Springer-Verlag, Berlin et al., 1993.Google Scholar
  14. 14.
    A.D. Jepson and D.J. Heeger. Subspace methods for recovering rigid motion II: Theory. Technical Report RBCV-TR-90-36, University of Toronto, 1990.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1996

Authors and Affiliations

  • Konstantinos Daniilidis
    • 1
    • 3
  • Inigo Thomas
    • 2
  1. 1.Computer Science InstituteUniversity of KielGermany
  2. 2.GRASP LaboratoryUniversity of PennsylvaniaUSA
  3. 3.KielGermany

Personalised recommendations