Vision by Graph Pyramids

  • Walter G. Kropatsch


To efficiently process huge amounts of structured sensory data for vision, graph pyramids are proposed. Hierarchies of graphs can be generated by dual graph contraction. The goal is to reduce the data structure by a constant reduction factor while preserving certain image properties, like connectivity. While implemented versions solve several technical vision problems like image segmentation, the framework can be used as a model for biological systems, too.


Receptive Field Boundary Segment Dual Graph Sensor Arrangement Primal Graph 
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. Sister M, Cornelis J, Rosenfeld A (1990) A critical view of pyramid segmentation algorithms. Pattern Recognition Letters, 11(No. 9), pp 605–617CrossRefGoogle Scholar
  2. Burge M, Kropatsch WG (1999) A minimal line property preserving representation of line images. Computing, Devoted Issue on Image Processing, 62, pp 355–368Google Scholar
  3. Burt PJ, Adelson EH (1983) The Laplacian pyramid as a compact image code. IEEE Transactions on Communications Vol. COM-31(No.4), pp 532–540Google Scholar
  4. Gareth AJ, Singerman D (1978) Theory of Maps on Orientable Surfaces. Vol. 3. London Mathematical SocietyGoogle Scholar
  5. Glantz R, Englert R, Kropatsch WG (1999) Representation of image structure by a pair of dual graphs. In: Kropatsch WG, Jolion J-M (eds), 2nd IAPR-TC-15 workshop on graph-based representation. österreichische Computer Gesellschaft, OCG-Schriftenreihe Band 126, pp 155–163Google Scholar
  6. Jolion J-M, Kropatsch WG (eds) (1998) Graph Based Representations in Pattern Recognition. Computing, Suppl 12. Springer-Verlag, Wien New YorkGoogle Scholar
  7. Jolion J-M, Rosenfeld A (1994) A Pyramid Framework for Early Vision. Kluwer Acad Publ DordrechtGoogle Scholar
  8. Jolion J-M, Kropatsch WG, Vento Mario (eds) (2001) Graph-based Representations in Pattern Recognition, GbR 2001. CUEN. ISBN 88 7146 579–582Google Scholar
  9. Kropatsch WG (1994) Building irregular pyramids by dual graph contraction. Tech rep PRIP-TR-35. Institute f. Automation 183/2, Pattern Recognition and Image Processing Group, TU Wien, Austria. Also available through Scholar
  10. Kropatsch WG (1997a) Equivalent contraction kernels to build dual irregular pyramids. Advances in Computer Vision, pp 99–107, Springer series Advances in Computer ScienceGoogle Scholar
  11. Kropatsch WG (1997b) Property preserving hierarchical graph transformations. In: Arcelli C, Cordella LP, Sanniti di Baja G (eds) Advances in Visual Form Analysis. World Scientific Publishing Company, pp 340–349Google Scholar
  12. Kropatsch WG (1999) How useful is structure in motion? In: Chetverikov D, Szirányi T (eds) Fundamental Structural Properties in Image and Pattern Analysis 1999. österreichische Computer Gesellschaft, OCG-Schriftenreihe Band 130, pp 35–45Google Scholar
  13. Kropatsch WG, BenYacoub S (1996) Universal segmentation with plRRamids. In: Pinz A (ed) Pattern Recognition 1996, Proc of 20th ÖAGM workshop R. Oldenburg, OCG-Schriftenreihe, Band 90, pp 171–182Google Scholar
  14. Kropatsch WG, Brun L (2000) Hierarchies of combinatorial maps. In: Svoboda T (ed) CPRW2000, Proceedings of the Czech Pattern Recognition Workshop. Perslák, CZ: Czech Pattern Recognition Society Praha. ISBN 80238–5215–9, pp 131–137Google Scholar
  15. Kropatsch WG, Jolion J-M (eds) (1999) 2nd IAPR-TC-15 Workshop on Graph-based Representation. Österreichische Computer Gesellschaft Band 126Google Scholar
  16. Lienhardt P (1989) Subdivisions of n-dimensional spaces and n-dimensional generalized maps. In: Mehlhorn K (ed) Proceedings of the 5th Annual Symposium on Computational Geometry (SCG ’89). Saarbrücken, FRG: ACM Press, pp 228–236CrossRefGoogle Scholar
  17. Macho H, Kropatsch WG (1995) Finding connected components with dual irregular pyramids. In: Solina F, Kropatsch WG (eds) Visual Modules, Proc of 19th ÖAGM and 1st SDVR workshop R. Oldenburg, OCG-Schriftenreihe Band 81, pp 313–321Google Scholar
  18. Marr D (1982) Vision. WH Freeman, San FranciscoGoogle Scholar
  19. Pailloncy J-G, Kropatsch WG, Jolion J-M (1998) Object matching on irregular pyramid. In: Jain AK, Venkatesh S, Lovell BC (eds) 14th International Conference on Pattern Recognition, Vol. II. IEEE Comp Soc, pp 1721–1723Google Scholar
  20. Pizlo Z, Salach-Golyska M, Rosenfeld A (1997) Curve detection in a noisy image. Vision Res 37(9): 1217–1241PubMedCrossRefGoogle Scholar
  21. Rosenfeld A, Nakamura A (1997) Local deformations of digital curves. Pattern Recogn Lett 18: 613–620CrossRefGoogle Scholar
  22. Thulasiraman K, Swamy MNS (1992) Graphs: Theory and Algorithms. Wiley-Interscience, New York, USAGoogle Scholar
  23. Uhr L, Schmitt L (1984) The several steps from icon to symbol using structured cone/pyramids. In: Rosenfeld A (ed) Multiresolution Image Processing and Analysis. Springer Verlag, Berlin Heidelberg New York Tokyo, pp 86–100CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Wien 2003

Authors and Affiliations

  • Walter G. Kropatsch

There are no affiliations available

Personalised recommendations