Scalable geometric hashing on MasPar machines

Ashfaq A. Khokhar, Viktor K. Prasanna, Hyong Joong Kim

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

In this paper, we present scalable data parallel algorithms for geometric hashing. We perform implementations of the proposed algorithms on MasPar MP-1/MP-2. In earlier parallel implementations, the number of processors employed is independent of the size of the scene but depends on the size of the model database which is usually very large. We design new parallel algorithms and map them onto MP-1/MP-2. These techniques significantly improve upon the number of processors employed while achieving superior time performance. Our implementations run on a P processor machine, such that 1 ≤ P ≤ S, where S is the number of feature points in the scene. Our results show that a probe of the recognition phase for a scene consisting of 1024 feature points takes less than 50 m sec on a 1K processor MP-1/MP-2.

Original languageEnglish
Title of host publicationIEEE Computer Vision and Pattern Recognition
Editors Anon
PublisherPubl by IEEE
Pages594-595
Number of pages2
ISBN (Print)0818638826
Publication statusPublished - 1993
Externally publishedYes
EventProceedings of the 1993 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - New York, NY, USA
Duration: 1993 Jun 151993 Jun 18

Other

OtherProceedings of the 1993 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
CityNew York, NY, USA
Period93/6/1593/6/18

Fingerprint

Parallel algorithms

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Khokhar, A. A., Prasanna, V. K., & Kim, H. J. (1993). Scalable geometric hashing on MasPar machines. In Anon (Ed.), IEEE Computer Vision and Pattern Recognition (pp. 594-595). Publ by IEEE.

Scalable geometric hashing on MasPar machines. / Khokhar, Ashfaq A.; Prasanna, Viktor K.; Kim, Hyong Joong.

IEEE Computer Vision and Pattern Recognition. ed. / Anon. Publ by IEEE, 1993. p. 594-595.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Khokhar, AA, Prasanna, VK & Kim, HJ 1993, Scalable geometric hashing on MasPar machines. in Anon (ed.), IEEE Computer Vision and Pattern Recognition. Publ by IEEE, pp. 594-595, Proceedings of the 1993 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, New York, NY, USA, 93/6/15.
Khokhar AA, Prasanna VK, Kim HJ. Scalable geometric hashing on MasPar machines. In Anon, editor, IEEE Computer Vision and Pattern Recognition. Publ by IEEE. 1993. p. 594-595
Khokhar, Ashfaq A. ; Prasanna, Viktor K. ; Kim, Hyong Joong. / Scalable geometric hashing on MasPar machines. IEEE Computer Vision and Pattern Recognition. editor / Anon. Publ by IEEE, 1993. pp. 594-595
@inproceedings{10f326fcefc94262b0199dc6c37bf74a,
title = "Scalable geometric hashing on MasPar machines",
abstract = "In this paper, we present scalable data parallel algorithms for geometric hashing. We perform implementations of the proposed algorithms on MasPar MP-1/MP-2. In earlier parallel implementations, the number of processors employed is independent of the size of the scene but depends on the size of the model database which is usually very large. We design new parallel algorithms and map them onto MP-1/MP-2. These techniques significantly improve upon the number of processors employed while achieving superior time performance. Our implementations run on a P processor machine, such that 1 ≤ P ≤ S, where S is the number of feature points in the scene. Our results show that a probe of the recognition phase for a scene consisting of 1024 feature points takes less than 50 m sec on a 1K processor MP-1/MP-2.",
author = "Khokhar, {Ashfaq A.} and Prasanna, {Viktor K.} and Kim, {Hyong Joong}",
year = "1993",
language = "English",
isbn = "0818638826",
pages = "594--595",
editor = "Anon",
booktitle = "IEEE Computer Vision and Pattern Recognition",
publisher = "Publ by IEEE",

}

TY - GEN

T1 - Scalable geometric hashing on MasPar machines

AU - Khokhar, Ashfaq A.

AU - Prasanna, Viktor K.

AU - Kim, Hyong Joong

PY - 1993

Y1 - 1993

N2 - In this paper, we present scalable data parallel algorithms for geometric hashing. We perform implementations of the proposed algorithms on MasPar MP-1/MP-2. In earlier parallel implementations, the number of processors employed is independent of the size of the scene but depends on the size of the model database which is usually very large. We design new parallel algorithms and map them onto MP-1/MP-2. These techniques significantly improve upon the number of processors employed while achieving superior time performance. Our implementations run on a P processor machine, such that 1 ≤ P ≤ S, where S is the number of feature points in the scene. Our results show that a probe of the recognition phase for a scene consisting of 1024 feature points takes less than 50 m sec on a 1K processor MP-1/MP-2.

AB - In this paper, we present scalable data parallel algorithms for geometric hashing. We perform implementations of the proposed algorithms on MasPar MP-1/MP-2. In earlier parallel implementations, the number of processors employed is independent of the size of the scene but depends on the size of the model database which is usually very large. We design new parallel algorithms and map them onto MP-1/MP-2. These techniques significantly improve upon the number of processors employed while achieving superior time performance. Our implementations run on a P processor machine, such that 1 ≤ P ≤ S, where S is the number of feature points in the scene. Our results show that a probe of the recognition phase for a scene consisting of 1024 feature points takes less than 50 m sec on a 1K processor MP-1/MP-2.

UR - http://www.scopus.com/inward/record.url?scp=0027835825&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027835825&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0027835825

SN - 0818638826

SP - 594

EP - 595

BT - IEEE Computer Vision and Pattern Recognition

A2 - Anon, null

PB - Publ by IEEE

ER -