Visual Form pp 495-506 | Cite as

Occulsions and Perspective in an Image Sequence

  • Amir Shmuel
  • Michael Werman

Abstract

This paper proposes an active solution to optimally recovering 3D scene depth from a sequence of pictures. Active vision, as referred to in this paper, is characterized by gaze control for input dependent data acquisition, coupled with a treatment of the reliability of the acquired information.

An active choice is described for choosing the difference between camera locations for the different pictures in the sequence. The trade-off between short and long distances between successive locations of the camera is shown in terms of occlusions, perspective transformation and exact triangulation.

The analysis also gives insight into the results and pitfalls of ordinary stereo.

Keywords

Optical Flow Camera Motion Stereo Image Active Vision Pixel Location 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    A.L. Abbot and N. Ahuja. Surface reconstruction by dynamic integration of focus, camera vergence and stereo. In Second International Conference on Computer Vision, pages 532-545, 1988.Google Scholar
  2. [2]
    J. Aloimonos, I. Weiss, and A. Bandyopadhyay. Active vision. In First International Conference on Computer Vision, pages 35-54, 1987.Google Scholar
  3. [3]
    D.H. Ballard and A. Ozcandarli. Eye fixation and early vision: kinetic depth. In Second International Conference on Computer Vision, pages 524-531, 1988.Google Scholar
  4. [4]
    J.J. Clark and N.J. Ferrier. Modal control of an attentive vision system. In Second International Conference on Computer Vision, pages 514-523, 1988.Google Scholar
  5. [5]
    M. A. Gennert. A Computational Framework for Understanding Problems in Stereo Vision. PhD thesis, Massachusetts Institute of Technology, 1987.Google Scholar
  6. [6]
    B. Horn. Robot Vision. The MIT Press, 1986.Google Scholar
  7. [7]
    D. Marr. Vision: A Computational Investigation into the Human Representation and Processing of Visual Information. Springer, 1987.Google Scholar
  8. [8]
    P. A. Ruymgaart and T. T. Soong. Mathematics of Kalman-Bucy Filtering. Springer, 1987.Google Scholar
  9. [9]
    A. Shmuel and M. Werman. Active vision: 3d depth from an image sequence. In Israeli Symposium AI, Vision and Patttern Recognition, pages 583-600, Tel Aviv, 1989.Google Scholar
  10. [10]
    A. Shmuel and M. Werman. Active vision: 3d from an image sequence. In 10’th International Conference on Pattern Recognition, Atlantic City, 1990.Google Scholar
  11. [11]
    M.A. Snyder. Uncertainty analysis of image measurements. In Image Understanding Workshop, pages 681-693, 1987.Google Scholar

Copyright information

© Springer Science+Business Media New York 1992

Authors and Affiliations

  • Amir Shmuel
    • 1
  • Michael Werman
    • 1
  1. 1.Department of Computer ScienceThe Hebrew University of JerusalemJerusalemIsrael

Personalised recommendations