Complexity of indexing: Efficient and learnable large database indexing

  • Michael Werman
  • Daphna Weinshall
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1064)


Object recognition starts from a set of image measurements (including locations of points, lines, surfaces, color, and shading), which provides access into a database where representations of objects are stored. We describe a complexity theory of indexing, a meta-analysis which identifies the best set of measurements (up to algebraic transformations) such that: (1) the representation of objects are linear subspaces and thus easy to learn; (2) direct indexing is efficient since the linear subspaces are of minimal rank. The index complexity is determined via a simple process, equivalent to computing the rank of a matrix. We readily re-derive the index complexity of the few previously analyzed cases. We then compute the best index for new cases: 6 points in one perspective image, and 6 directions in one para-perspective image; the most efficient representation of a color is a plane in 3D space. For future applications with any vision problem where the relations between shape and image measurements can be written down in an algebraic form, we give an automatic process to construct the most efficient database that can be directly obtained by learning from examples.


Image Measurement Efficient Index Vision Problem Invariant Relation Class Constraint 
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.
    D. T. Clemens and D. W. Jacobs. Space and time bounds on indexing 3-D models from 2-D images. IEEE T-PAMI, 13(10):1007–1017, 1991.Google Scholar
  2. 2.
    D. P. Huttenlocher and S. Ullman. Recognizing solid objects by alignment with an image. IJCV, 5:195–212, 1990.Google Scholar
  3. 3.
    D. W. Jacobs. Matching 3-D models to 2-D images. IJCV, 1995. in press.Google Scholar
  4. 4.
    Y. Lamdan and H. Wolfson. Geometric hashing: a general and efficient recognition scheme. In Proc. 2nd ICCV, pages 238–251, Tarpon Springs, FL, 1988. IEEE, Washington, DC.Google Scholar
  5. 5.
    L. T. Maloney and B. Wandell. A computational model of color constancy. JOSA, 1:29–33, 1986.Google Scholar
  6. 6.
    Y. Moses and S. Ullman. Limitations of non model-based recognition schemes. In G. Sandini, editor, Proc. 2nd ECCV, Lecture Notes in Computer Science, volume 588, pages 820–828. Springer Verlag, 1992.Google Scholar
  7. 7.
    H. Murase and S. K. Nayar. Visual learning and recognition of 3-D objects from appearance. IJCV, 1995. in press.Google Scholar
  8. 8.
    C.A. Rothwell, D.A. Forsyth, A. Zisserman, and J.L. Mundy. Extracting projective structure from single perspective views of 3d point sets. In Proc. 4th ICCV, pages 573–582, Berlin, Germany, 1993. IEEE, Washington, DC.Google Scholar
  9. 9.
    D. Weinshall. Model-based invariants for 3D vision. IJCV, 10(1):27–42, 1993.Google Scholar
  10. 10.
    D. Weinshall, M. Werman, and A. Shashua. Duality of multi-point and multiframe Geometry: Fundamental Shape Matrices and Tensors. In Proc. 4th ECCV, Cambridge, UK, 1996.Google Scholar
  11. 11.
    M. Werman and A. Shashua. Elimination: An approach to the study of 3D-from-2D. In Proc. ICCV, June 1995.Google Scholar
  12. 12.
    M. Werman and D. Weinshall. Complexity of indexing: Efficient and learnable large database indexing. TR 95–7, Hebrew University, 1995.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1996

Authors and Affiliations

  • Michael Werman
    • 1
  • Daphna Weinshall
    • 1
  1. 1.Institute of Computer ScienceThe Hebrew University of JerusalemJerusalemIsrael

Personalised recommendations