Using two-dimensional models to interact with the three-dimensional world

  • Daniel P. Huttenlocher
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 994)


This paper describes a method for tracking a moving object in an image, when the camera motion is unknown and other moving objects may be in the image. The method is based on matching two-dimensional geometric structures between successive frames of an image sequence. A bitmap representing the object being tracked at one time frame is matched to features extracted from the image at the next time frame. The transformation mapping the object to the image specifies a new model of the object for the subsequent frame. The approach makes no use of optical flow estimates nor of three-dimensional information. We present examples of the method for tracking moving objects in video sequences and for visual guidance of a mobile robot.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    P. Bouthemy and E. Francois, Motion Segmentation and Qualitative Dynamic Scene Analysis from an Image Sequence, Intl. Journal of Computer Vision, 10(2), pp. 157–182, 1993.Google Scholar
  2. 2.
    J.F. Canny. A computational approach to edge detection. IEEE Trans. Pat. Anal. and Mach. Intel., 8(6):34–43, 1986.Google Scholar
  3. 3.
    E. Dickmanns and V. Graefe. Dynamic monocular machine vision. Machine Vision Applications, 1:223–240, 1988.Google Scholar
  4. 4.
    D. Gennery. Tracking known three dimensional objects. In Second National Conf. on Artificial Intelligence, pages 13–17, 1982.Google Scholar
  5. 5.
    D.P. Huttenlocher, G.A. Klanderman and W.J. Rucklidge, Comparing Images Using the Hausdorff Distance, IEEE Trans. Pat. Anal. and Mach. Intel., 15(9), pp. 850–863, 1993.Google Scholar
  6. 6.
    D.P. Huttenlocher and W.J. Rucklidge, A Multi-Resolution Technique for Comparing Images Using the Hausdorff Distance, Proceedings of the IEEE Computer Vision and Pattern Recognition Conference, pp. 705–706, 1993.Google Scholar
  7. 7.
    M. Kass, A. Witkin, and D. Terzopoulos. Snakes: Active contour models. International Journal of Computer Vision, 1(3):321–331, 1988.CrossRefGoogle Scholar
  8. 8.
    D. Koller, K. Daniilidis, T. Thórhallsson, and H.-H. Nagel. Model-based object tracking in traffic scenes. In Proc. 2nd European Conf. on Computer Vision, pages 437–452, Santa Margherita Ligure, Italy, May 1992.Google Scholar
  9. 9.
    D.G. Lowe. Robust model-based motion tracking through the integration of search and estimation. International Journal of Computer Vision, 8(2):113–122, 1992.Google Scholar
  10. 10.
    N. Ueda and K. Mase. Tracking moving contours using energy-minimizing elastic contour models. In Proc. 2nd European Conf. on Computer Vision, pages 453–457, Santa Margherita Ligure, Italy, May 1992.Google Scholar
  11. 11.
    W.J. Rucklidge, Efficient Computation of the Minimum Hausdorff Distance for Visual Recognition, PhD thesis, Computer Science Dept., Cornell Univ., 1995.Google Scholar
  12. 12.
    W.J. Rucklidge, Locating Objects Using the Hausdorff Distance, Proc. Fifth Intnl. Conf. on Computer Vision, 1995.Google Scholar
  13. 13.
    G. Verghese, K. Gale, and C.R. Dyer. Real-time, parallel motion tracking of three-dimensional objects from spatiotemporal image sequences. In Kumar et. al., editor, Parallel Algorithms for Machine Intelligence and Vision, pages 340–359, New York, 1990. Springer-Verlag.Google Scholar
  14. 14.
    J. Woodfill and R.D. Zabih. An algorithm for real-time tracking of non-rigid objects. In Proc. American Association for Artificial Intelligence Conference, 1991.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1995

Authors and Affiliations

  • Daniel P. Huttenlocher
    • 1
    • 2
  1. 1.Computer Science DepartmentCornell UniversityUSA
  2. 2.Palo Alto Research CenterXeroxUSA

Personalised recommendations