Visual Form pp 623-632 | Cite as

Invariants and Pose Determination

  • Xinming Yu
  • T. D. Bui
  • A. Krzyżak


A quadratic surface can be expressed with 3 scalar invariants as shape parameter and a transformation matrix as pose parameter. Surface representation is insensitive to partial occlusions. Quadratic surface patch correspondence by their invariants can be performed very efficiently (in O(log2 n) time, where n is the number of quadratic surface patches in one object). After matching of patches, the position and orientation between the two matched patches can be determined by the pose matrices and used as rigidity constraint for object recognition. Various degenerate cases are analyzed. Experimental results are provided using synthetic range data with Gaussian noise.


Object Recognition Transformation Matrix Object Model Rotation Matrix Degenerate Case 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    J. (Yiannis) Aloimonos and I. Rigoutsos. Determining the 3-D Motion of a Rigid Surface Patch Without Correspondence, Under Perspective Projection: i. Planar Surfaces, ii. Curved Surfaces. In Proc. aaai86, pp. 681-688, Philadelphia, PA, 1986.Google Scholar
  2. [2]
    K. S. Arun, T. S. Huang, and S. D. Blostein. Least-squares Fitting of Two 3-D Point Sets. IEEE PAMI-9(5), Sept. 1987.Google Scholar
  3. [3]
    Fred L. Bookstein. Fitting Conic Sections to Scattered Data. Computer Graphics and Image Processing, 9:56–71, 1979.CrossRefGoogle Scholar
  4. [4]
    O. D. Faugeras and M. Hebert. The Representation, Recognition, and Locating of 3-D Objects. The International Journal of Robotics Research, 5(3):27–52, Fall 1986.CrossRefGoogle Scholar
  5. [5]
    F. P. Ferrie. Reconstructing and Interpreting the 3D Shape of Moving Objects. PhD thesis, McGill University, Montreal, Quebec, Canada, March 1986.Google Scholar
  6. [6]
    D. Forsyth, J. L. Mundy, A. Zisserman, and C. M. Brown. Projectively Invariant Representations using Implicit Algebraic Curves. In Proc. of the First European Conf. on Comp. Vision — ECCV90, pp. 427-436, April 1990. Springer-Verlag.Google Scholar
  7. [7]
    R. Gonin and A. H. Money. Nonlinear L p-Norm Estimation. Marcel Dekker, Inc., New York, 1989.Google Scholar
  8. [8]
    E. L. Hall, J. B. K. Tio, C. A. McPherson, and F. A. Sadjadi. Measuring Curved Surfaces for Robot Vision. IEEE Computer, 15(12):42–54, 1982.CrossRefGoogle Scholar
  9. [9]
    G. A. Korn and T. M. Korn. Mathematical Handbook for Scientists and Engineers. McGraw-Hill, New York, N.Y., 1961.zbMATHGoogle Scholar
  10. [10]
    Z. C. Lin, H. Lee, and T. S. Huang. Finding 3-D Point Correspondences in Motion Estimation. In Proc. of the 8th Int. Conf. on Pattern Recog., Oct. 1986.Google Scholar
  11. [11]
    C. H. Lo and H. S. Don. 3-D Moment Forms: Their Construction and Application to Object Identification and Positioning. IEEE PAMI-11(10):1053–1064, Oct. 1989.Google Scholar
  12. [12]
    Xinming Yu, T. D. Bui, and A. Krzyzȧk. Robust Contour Decomposition by Residual Consensus. To appear in Proceedings of the Canadian Conference on Electrical and Computer Engineering, Québec, Canada, Sept 25-27 1991.Google Scholar

Copyright information

© Springer Science+Business Media New York 1992

Authors and Affiliations

  • Xinming Yu
    • 1
  • T. D. Bui
    • 1
  • A. Krzyżak
    • 1
  1. 1.Computer Science DepartmentConcordia UniversityMontréalCanada

Personalised recommendations