Parallelizing Vision Computations on CM-5: Algorithms and Experiences

  • Viktor K. Prasanna
  • Cho-Li Wang


This chapter summarizes our work in using Connection Machine CM-5 for vision. We define a realistic model of CM-5 in which explicit cost is associated with data routing and cooperative operations. Using this model, we develop scalable parallel algorithms for representative problems in vision computations at all three levels: low-level, intermediate-level and high-level.


Hash Function Hash Table Image Block Perceptual Group Total Execution Time 
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]
    A. Aho, J. Hoperoft, and J. Ullman, Data Structures and Algorithms, Addision-Wesley, 1983.Google Scholar
  2. [2]
    O. Bourdon and G. Medioni, “Object Recognition Using Geometric Hashing on the Connection Machine,” International Conference on Pat. tern Recognition, pages 596–600, 1988.Google Scholar
  3. [3]
    J. G. Dunham, “Optimum Uniform Piecewise Linear Approximation of Planar Curves,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 8, No. 1, pages 67–75, 1986.CrossRefGoogle Scholar
  4. [4]
    A. Huertas, C. Lin, and R. Nevatia, “Detection of Buildings from Mono-cular Views of Aerial Scenes Using Perceptual Grouping and Shadows,” Image Understanding Workshop, pages 253–260, 1993.Google Scholar
  5. [5]
    V. Kumar, A. Grama, A. Gupta, and G. Karypis, Introduction to Parallel Computing: Design and Analysis of Parallel Algorithms, Benjamin/Cummings, 1994.Google Scholar
  6. [6]
    T. Kwan, B. Totty, and D. Reed, “Communication and Computation Performance of the CM-5,” Proc. of Supercomputing ‘83, pages 192–201, 1993.Google Scholar
  7. [7]
    C. Leiserson, “The Network Architecture of Connection Machine CM-5,” Technical Report, Thinking Machines Corporation, 1992.Google Scholar
  8. [8]
    F. Leighton, “Tight Bounds on the Complexity of Parallel Sorting,” IEEE Transactions on Computers, Vol. 34, No. 4, pages 344–354, 1985.MathSciNetzbMATHCrossRefGoogle Scholar
  9. [9]
    R. Nevatia and K. Babu, “Linear Feature Extraction and Description,” Computer Graphics and Image processing, Vol. 13, pages 257–269, 1980.Google Scholar
  10. [10]
    V. Prasanna and C. Wang, “Scalable Parallel Implementations of Perceptual Grouping on Connection Machine CM-5,” in International Conference on Pattern Recognition, 1994.Google Scholar
  11. [11]
    V. Prasanna and C. Wang, “Image Feature Extraction on Connection Machine CM-5,” in Image Understanding Workshop, pages 595–602, 1994.Google Scholar
  12. [12]
    I. Rogoutsos and R. Hummel, “Massively Parallel Model Matching: Geometric Hashing on the Connection Machine,” IEEE Computer, pages 33–42, 1992.Google Scholar
  13. [13]
    J. Roberge, “A Data Reduction Algorithm for Planar Curves,” Computer Vision, Graphics, and Image Processing, Vol. 29, pages 168–195, 1985.zbMATHGoogle Scholar
  14. [14]
    Thinking Machines Corporation, CMMD Reference Guide Version 3. 0, 1992.Google Scholar
  15. [15]
    C. Wang, V. K. Prasanna, H. Kim, and A. Khokhar, “Scalable Data Parallel Implementations of Object Recognition using Geometric Hashing,” Journal of Parallel and Distributed Computing, pages 96–109, March 1994.Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 1995

Authors and Affiliations

  • Viktor K. Prasanna
    • 1
  • Cho-Li Wang
    • 1
  1. 1.Department of EE-SystemsUniversity of Southern CaliforniaLos AngelesUSA

Personalised recommendations