Matching perspective views of parallel plane structures

  • Luc Van Gool
  • Theo Moons
  • Marc Van Diest
  • Eric Pauwels
Part of the Lecture Notes in Computer Science book series (LNCS, volume 825)


Within an invariance framework, the recognition of plane objects under general viewpoints and perspective projection calls for the extraction of two-dimensional projective invariants. If the possible poses of the object are constrained with respect to the camera, however, simpler groups than the projective transformations become relevant, and consequently, simpler invariants exist. Several such special types of pose constraints are discussed — all amount to the object plane remaining parallel to its original orientation — and the corresponding groups are outlined. For each group a number of invariants are derived to illustrate the gain in simplicity.


Reference Point Object Plane Projective Transformation Perspective Projection General Translation 
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. [BBP1]
    M. Brill, E. Barrett, and P. Payton, Projective invariants for curves in two and three dimensions, in Geometric Invariance in Computer Vision, eds. Mundy & Zisserman, pp. 193–214, MIT Press, 1992.Google Scholar
  2. [CB1]
    R. Collins and J. Beveridge, Matching perspective views of coplanar structures using projective unwarping and similarity matching, Conf. Computer Vision Pattern Recognition, pp. 240–245, 1992.Google Scholar
  3. [LTD1]
    R. Lotufo, B. Thomas, and E. Dagless, Road following algorithm using a panned plan-view transformation, Proc. 1st ECCV, pp. 231–235, 1990.Google Scholar
  4. [MZB1]
    D. Mukherjee, A. Zisserman, and M. Brady, Shape from symmetry — detecting and exploiting symmetry in affine images, Techn. Report Univ. of Oxford, OUEL 1988/93, 1993.Google Scholar
  5. [SCC1]
    M. Straforini, C. Coelho, and M. Campani, Extraction of vanishing points from images of indoor and outdoor scenes, Image and Vision Computing, Vol. 11, no.2, pp. 91–99, march 1993.Google Scholar
  6. [Tea1]
    A. Tai, J. Kittler, M. Petrou, and T. Windeatt, Vanishing point detection, Image and Vision Computing, Vol. 11, No.4, pp. 240–245, 1993.Google Scholar
  7. [TBS1]
    T. Tan, K. Baker, and G. Sullivan, 3D structure and motion estimation from 2D image sequences, Image and Vision Computing, vol. 11, no.4, pp. 203–210, 1993.Google Scholar
  8. [VKO1]
    L. Van Gool, P. Kempenaers. and A. Oosterlinck, Recognition and semi-differential invariants, Proc. CVPR, pp. 454–460, june 1991Google Scholar
  9. [Vea1]
    L. Van Gool, T. Moons, E. Pauwels, and A. Oosterlinck, Semi-differential invariants, in Geometric Invariance in Computer Vision, eds. Mundy & Zisserman, pp. 157–192, MIT Press, 1992.Google Scholar
  10. [Vea2]
    L. Van Gool, T. Moons, D. Ungureanu, and A. Oosterlinck, The characterization and detection of skewed symmetry, Kath. Univ. Leuven, Techn. Report KUL/ESAT/MI2/9304, 1993, accepted for publication in CVGIP:IU.Google Scholar
  11. [Vea3]
    L. Van Gool, T. Moons, E. Pauwels, and J. Wagemans, Invariance from the Euclidean geometer's perspective, accepted for publication in Perception.Google Scholar
  12. [Vea4]
    L. Van Gool, T. Moons, M. Van Diest, and E. Pauwels, Perspective matching and tracking of moving plane structures with constant object plane orientation, Kath. Univ. Leuven, Techn. Report KUL/ESAT/MI2/9305, 1993.Google Scholar
  13. [Zea1]
    T. Zielke, K. Storjohann, H. Mallot, and W. von Seelen, Adaptive computer vision systems to the visual environment: topographic mapping, Proc. 1st ECCV 90, pp. 613–615, 1990.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1994

Authors and Affiliations

  • Luc Van Gool
    • 1
  • Theo Moons
    • 1
  • Marc Van Diest
    • 1
  • Eric Pauwels
    • 1
  1. 1.Katholieke Universiteit Leuven, ESAT-MI2LeuvenBelgium

Personalised recommendations