• Mapping stereo and image matching algorithms onto parallel architectures

    • Fulltext


        Click here to view fulltext PDF

      Permanent link:

    • Keywords


      Stereo matching; image matching; linear features; parallel algorithms; fixed size arrays; object recognition; shape from depth

    • Abstract


      In this paper we present parallel implementations of two vision tasks; stereo matching and image matching. Linear features are used as matching primitives. These implementations are performed on a fixed size mesh array and achieve processor-time optimal performance. For stereo matching, we proposeO(Nn3/P2) time algorithm on aP ×P processor mesh array, whereN is the number of line segments in one image,n is the number of line segments in a window determined by the object size, andPn. The sequential algorithm takesO(Nn3) time. For image matching, a partitioned parallel implementation is developed.O[((nm/P2) +P)nm] time performance is achieved on aP ×P processor mesh array, whereP2nm. This leads to a processor-time optimal solution forP ⩽ (nm)1/3.

    • Author Affiliations


      Ashfaq Khokhar1 Viktor K Prasanna1

      1. Department of EE-Systems, EEB 244, University of Southern California, Los Angeles, CA - 90089-2562, USA
    • Dates

  • Sadhana | News

    • Editorial Note on Continuous Article Publication

      Posted on July 25, 2019

      Click here for Editorial Note on CAP Mode

© 2021-2022 Indian Academy of Sciences, Bengaluru.