Robust line-based pose enumeration from A single image

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

6 Citations (Scopus)

Abstract

This paper proposes robust algorithms for line-based pose enumeration from a single view, and it reports on their evaluations by simulations. The proposed algorithms incorporate two major refinements into the algorithms originally proposed by Shakunaga. The first refinement, introduction of zone-crossing detection to the 1-d search remarkably decreases the rate of overlooking a correct pose. The second refinement, adaptive selection of a PAT pair considerably reduces the average estimation error. Simulation results show that pose estimation precision depends primarily on the precision of line detection. Although the refinements are widely effective, they are more effective for more precise line detection. For 99% of rigid body samples, the algorithm can estimate rotation with an error of less than 2 degrees, and for 99.9% of the samples, the error is less than 10 degrees. Simulation experiments for articulated objects show similar results by using the second algorithm. The effectiveness of the algorithms is verified in an alignment approach by simulations and by real processings.

Original languageEnglish
Title of host publication1993 IEEE 4th International Conference on Computer Vision
PublisherPubl by IEEE
Pages545-550
Number of pages6
ISBN (Print)0818638729
Publication statusPublished - 1993
Externally publishedYes
Event1993 IEEE 4th International Conference on Computer Vision - Berlin, Ger
Duration: May 11 1993May 14 1993

Other

Other1993 IEEE 4th International Conference on Computer Vision
CityBerlin, Ger
Period5/11/935/14/93

Fingerprint

Error analysis
Processing
Experiments

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Shakunaga, T. (1993). Robust line-based pose enumeration from A single image. In 1993 IEEE 4th International Conference on Computer Vision (pp. 545-550). Publ by IEEE.

Robust line-based pose enumeration from A single image. / Shakunaga, Takeshi.

1993 IEEE 4th International Conference on Computer Vision. Publ by IEEE, 1993. p. 545-550.

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

Shakunaga, T 1993, Robust line-based pose enumeration from A single image. in 1993 IEEE 4th International Conference on Computer Vision. Publ by IEEE, pp. 545-550, 1993 IEEE 4th International Conference on Computer Vision, Berlin, Ger, 5/11/93.
Shakunaga T. Robust line-based pose enumeration from A single image. In 1993 IEEE 4th International Conference on Computer Vision. Publ by IEEE. 1993. p. 545-550
Shakunaga, Takeshi. / Robust line-based pose enumeration from A single image. 1993 IEEE 4th International Conference on Computer Vision. Publ by IEEE, 1993. pp. 545-550
@inproceedings{91fc603dfc2c4c029f235b93af7687ec,
title = "Robust line-based pose enumeration from A single image",
abstract = "This paper proposes robust algorithms for line-based pose enumeration from a single view, and it reports on their evaluations by simulations. The proposed algorithms incorporate two major refinements into the algorithms originally proposed by Shakunaga. The first refinement, introduction of zone-crossing detection to the 1-d search remarkably decreases the rate of overlooking a correct pose. The second refinement, adaptive selection of a PAT pair considerably reduces the average estimation error. Simulation results show that pose estimation precision depends primarily on the precision of line detection. Although the refinements are widely effective, they are more effective for more precise line detection. For 99{\%} of rigid body samples, the algorithm can estimate rotation with an error of less than 2 degrees, and for 99.9{\%} of the samples, the error is less than 10 degrees. Simulation experiments for articulated objects show similar results by using the second algorithm. The effectiveness of the algorithms is verified in an alignment approach by simulations and by real processings.",
author = "Takeshi Shakunaga",
year = "1993",
language = "English",
isbn = "0818638729",
pages = "545--550",
booktitle = "1993 IEEE 4th International Conference on Computer Vision",
publisher = "Publ by IEEE",

}

TY - GEN

T1 - Robust line-based pose enumeration from A single image

AU - Shakunaga, Takeshi

PY - 1993

Y1 - 1993

N2 - This paper proposes robust algorithms for line-based pose enumeration from a single view, and it reports on their evaluations by simulations. The proposed algorithms incorporate two major refinements into the algorithms originally proposed by Shakunaga. The first refinement, introduction of zone-crossing detection to the 1-d search remarkably decreases the rate of overlooking a correct pose. The second refinement, adaptive selection of a PAT pair considerably reduces the average estimation error. Simulation results show that pose estimation precision depends primarily on the precision of line detection. Although the refinements are widely effective, they are more effective for more precise line detection. For 99% of rigid body samples, the algorithm can estimate rotation with an error of less than 2 degrees, and for 99.9% of the samples, the error is less than 10 degrees. Simulation experiments for articulated objects show similar results by using the second algorithm. The effectiveness of the algorithms is verified in an alignment approach by simulations and by real processings.

AB - This paper proposes robust algorithms for line-based pose enumeration from a single view, and it reports on their evaluations by simulations. The proposed algorithms incorporate two major refinements into the algorithms originally proposed by Shakunaga. The first refinement, introduction of zone-crossing detection to the 1-d search remarkably decreases the rate of overlooking a correct pose. The second refinement, adaptive selection of a PAT pair considerably reduces the average estimation error. Simulation results show that pose estimation precision depends primarily on the precision of line detection. Although the refinements are widely effective, they are more effective for more precise line detection. For 99% of rigid body samples, the algorithm can estimate rotation with an error of less than 2 degrees, and for 99.9% of the samples, the error is less than 10 degrees. Simulation experiments for articulated objects show similar results by using the second algorithm. The effectiveness of the algorithms is verified in an alignment approach by simulations and by real processings.

UR - http://www.scopus.com/inward/record.url?scp=0027150582&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027150582&partnerID=8YFLogxK

M3 - Conference contribution

SN - 0818638729

SP - 545

EP - 550

BT - 1993 IEEE 4th International Conference on Computer Vision

PB - Publ by IEEE

ER -