Active 3D object recognition using 3D affine invariants

  • Sven Vinther
  • Roberto Cipolla
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 801)


We evaluate the power of 3D affine invariants in an object recognition scheme. These invariants are actively estimated by Kalman filtering the data obtained from real-time tracking of image features through a sequence of images. Object information is stored and retrieved in a hash table using the invariants as stable indices. Recognition takes place when significant evidence for a particular shape has been found from the table. Results with real data are presented, and the noise problems arising due to the weak perspective approximation and corner localisation errors are discussed. Preliminary results for extending this method to multiple object recognition in cluttered scenes are also presented.


Kalman Filter Object Recognition Hash Table Multiple View Perceptual Grouping 
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.T. Chin and C.R. Dyer. Model-Based Recognition in Robot Vision. ACM Computing Surveys, 18(1):67–108, 1986.Google Scholar
  2. 2.
    D.G. Lowe. The viewpoint consistency constraint. Int. Journal of Computer Vision, 1:57–72, 1987.Google Scholar
  3. 3.
    D.P. Huttenlocher and S. Ullman. Recognising solid objects by alignment with an image. Int. Journal of Computer Vision, 5(2):195–212, 1990.Google Scholar
  4. 4.
    W.E.L. Grimson and T. Lozano-Perez. Localising Overlapping Parts by Searching the Interpretation Tree. IEEE Trans. Pattern Analysis and Machine Intell., 9(4):469–482, 1987.Google Scholar
  5. 5.
    J.L. Mundy and A.Zissermann editors. Geometric Invariance in Computer Vision. MIT Press, 1992.Google Scholar
  6. 6.
    L.G. Roberts. Machine perception of three — dimensional solids. In J.T. Tippet, editor, Optical and Electro-optical Information Processing. MIT Press, 1965.Google Scholar
  7. 7.
    J.J. Koenderink and A.J. van Doorn. Affine structure from motion. J. Opt. Soc. America, 8(2):377–385, 1991.Google Scholar
  8. 8.
    D. Weinshall. Model-Based Invariants for 3-D Vision. Int. Journal of Computer Vision, 10(1):27–42, 1993.Google Scholar
  9. 9.
    J. B. Burns, R. S. Weiss, and E. M. Riseman. View variation of point-set and line-segment features. IEEE Transactions on Pattern Analysis and Machine Intelligence, 15(1):51–68, January 1993.Google Scholar
  10. 10.
    R. Fawcett, A. Zisserman, and M. Brady. Extracting structure from an affine view of a 3D point set with one or two bilateral symmetries. In Proceedings of the British Machine Vision Conference, pages 349–358, Guildford, 1993.Google Scholar
  11. 11.
    S. Vinther and R. Cipolla. Towards 3D object model acquisition and recognition using 3D affine invariants. In Proc. British Machine Vision Conference 1993, pages 369–378, 1993.Google Scholar
  12. 12.
    S. Vinther and R. Cipolla. Active 3D object recognition using 3D affine invariants. Technical Report CUED/F-INFENG/TR164, Dept. of Engineering, University of Cambridge, 1994.Google Scholar
  13. 13.
    Y. Lamdan, J.T. Schwartz, and H.J. Wolfson. Affine Invariant Model-Based Object Recognition. IEEE Trans. on Robotics and Automation, 6(5):578–589, 1990.Google Scholar
  14. 14.
    W.E. Grimson, D.P. Huttenlocher, and D. Jacobs. A study of affine matching with bounded sensor error. In Proc. 2rd European Conf. on Computer Vision, 1992.Google Scholar
  15. 15.
    D.G. Lowe. Three dimensional object recognition from single two-dimensional images. Artificial Intelligence, 31:355–395, 1987.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1994

Authors and Affiliations

  • Sven Vinther
    • 1
  • Roberto Cipolla
    • 1
  1. 1.Department of EngineeringUniversity of CambridgeCambridgeEngland

Personalised recommendations