A practical algorithm for structure and motion recovery from long sequence of images

  • Miroslav Trajković
  • Mark Hedley
Poster Session B: Active Vision, Motion, Shape, Stereo
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1310)

Abstract

In this paper we present an algorithm for structure and motion (SM) recovery under affine projection from video sequences that is suitable for real time applications. The algorithm tracks the motion of a single structure, be it an object or the entire scene itself, allowing for any type of camera motion. This could be used for example to track the motion of a vehicle in a warehouse (single object, static camera) or for visual navigation from a moving platform (track scene from moving camera). The algorithm requires a set of features to be detected in each frame, and that at least four features are correctly matched between each three consecutive frames. Compared to previous algorithms, this novel algorithm has a lower computational cost, dynamically detects outliers and allows for previously lost features to reappear in the sequence. The algorithm has been tested on real image sequences, and compared to other algorithms we have found that our algorithm has both a smaller error and a lower computational time.

Keywords

Feature Point Singular Value Decomposition Motion Parameter Extend Kalman Filter Consecutive Frame 
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.

References

  1. [1]
    Held A., “Piecewise Shape Reconstruction by incremental factorization”, Proc.7 th British Machine Vision Conference, vol I, UK, 1996, pp. 333–342.Google Scholar
  2. [2]
    McLauchlan P. et al. “Recursive affine structure and motion from image sequences” Proc. 3 rd ECCV, vol. I, Stockholm, pp. 217–224, 1994.Google Scholar
  3. [3]
    McLauchlan P. and Murray D.. “A unifying framework for structure and motion recovery from image sequences” Proc. 5th ICCV, MIT, pp. 314–320, 1995.Google Scholar
  4. [4]
    Meer P. et al. “Robust regression methods for computer vision: a review”, Intl. Journal Computer Vision vol. 6., pp. 59–70, 1991.Google Scholar
  5. [5]
    Morita T. and Kanade T., “A sequential factorization method for recovering shape and motion from image streams”, Proc. 1994 ARPA Image Understanding Workshop, vol. II, pp. 1177–1188, Monterey, CA 1994.Google Scholar
  6. [6]
    Reid I. D. and Murray D. W., “Active Tracking of Foveated Feature Clusters Using Affine Structure”, Intl. J. Computer Vision, vol. 18, pp. 41–60, 1996.Google Scholar
  7. [7]
    Shi J. and Tomasi C., “Good Features to Track”. IEEE Conference on Computer Vision and Pattern Recognition, June 1994, pp. 593–600.Google Scholar
  8. [8]
    Tomasi C. and Kanade T., “Shape and motion from image streams under orthography: A factorization approach”, Intl. J. Comp. Vision, vol. 9, pp. 137–154, 1991.Google Scholar
  9. [9]
    Trajković M. and Hedley M., “Fast feature detection and matching for machine vision”, Proc. 7 th British Machine Vision Conference, vol. I, Edinburgh, UK, 1996, pp. 91–100.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • Miroslav Trajković
    • 1
  • Mark Hedley
    • 1
  1. 1.Department of Electrical EngineeringSydney UniversityAustralia

Personalised recommendations