Map estimation of epipolar geometry by em algorithm and local diffusion

Wenfeng Li, Baoxin Li

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

1 Citation (Scopus)

Abstract

Finding epipolar geometry for two images is a fundamental problem in computer vision. While this typically relies on feature point correspondence, the epipolar constraint can also be used for improving the accuracy of correspondence. We propose a probabilistic framework for estimating the epiploar geometry, in which the geometry and the feature correspondence are estimated iteratively at the same time. Using the EM algorithm to maximize a posteriori, our approach updates feature correspondence with estimated epipolar geometry. The correspondence is further improved with local diffusion on a prior Markov Random Field model. In turn, more accurate epipolar geometry is recovered. Experiments show this approach produces more accurate fundamental matrix compared with typical methods and can handle some challenging situations such as view rotation and scale changes.

Original languageEnglish (US)
Title of host publicationProceedings - International Conference on Image Processing, ICIP
Volume5
DOIs
StatePublished - 2006
Event14th IEEE International Conference on Image Processing, ICIP 2007 - San Antonio, TX, United States
Duration: Sep 16 2007Sep 19 2007

Other

Other14th IEEE International Conference on Image Processing, ICIP 2007
CountryUnited States
CitySan Antonio, TX
Period9/16/079/19/07

Fingerprint

Geometry
Computer vision
Experiments

Keywords

  • EM algorithm
  • Epipolar geometry
  • Local diffusion
  • MAP

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Li, W., & Li, B. (2006). Map estimation of epipolar geometry by em algorithm and local diffusion. In Proceedings - International Conference on Image Processing, ICIP (Vol. 5). [4379800] https://doi.org/10.1109/ICIP.2007.4379800

Map estimation of epipolar geometry by em algorithm and local diffusion. / Li, Wenfeng; Li, Baoxin.

Proceedings - International Conference on Image Processing, ICIP. Vol. 5 2006. 4379800.

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

Li, W & Li, B 2006, Map estimation of epipolar geometry by em algorithm and local diffusion. in Proceedings - International Conference on Image Processing, ICIP. vol. 5, 4379800, 14th IEEE International Conference on Image Processing, ICIP 2007, San Antonio, TX, United States, 9/16/07. https://doi.org/10.1109/ICIP.2007.4379800
Li W, Li B. Map estimation of epipolar geometry by em algorithm and local diffusion. In Proceedings - International Conference on Image Processing, ICIP. Vol. 5. 2006. 4379800 https://doi.org/10.1109/ICIP.2007.4379800
Li, Wenfeng ; Li, Baoxin. / Map estimation of epipolar geometry by em algorithm and local diffusion. Proceedings - International Conference on Image Processing, ICIP. Vol. 5 2006.
@inproceedings{7ce4f966ff554e2f8c53fc687deff0a0,
title = "Map estimation of epipolar geometry by em algorithm and local diffusion",
abstract = "Finding epipolar geometry for two images is a fundamental problem in computer vision. While this typically relies on feature point correspondence, the epipolar constraint can also be used for improving the accuracy of correspondence. We propose a probabilistic framework for estimating the epiploar geometry, in which the geometry and the feature correspondence are estimated iteratively at the same time. Using the EM algorithm to maximize a posteriori, our approach updates feature correspondence with estimated epipolar geometry. The correspondence is further improved with local diffusion on a prior Markov Random Field model. In turn, more accurate epipolar geometry is recovered. Experiments show this approach produces more accurate fundamental matrix compared with typical methods and can handle some challenging situations such as view rotation and scale changes.",
keywords = "EM algorithm, Epipolar geometry, Local diffusion, MAP",
author = "Wenfeng Li and Baoxin Li",
year = "2006",
doi = "10.1109/ICIP.2007.4379800",
language = "English (US)",
isbn = "1424414377",
volume = "5",
booktitle = "Proceedings - International Conference on Image Processing, ICIP",

}

TY - GEN

T1 - Map estimation of epipolar geometry by em algorithm and local diffusion

AU - Li, Wenfeng

AU - Li, Baoxin

PY - 2006

Y1 - 2006

N2 - Finding epipolar geometry for two images is a fundamental problem in computer vision. While this typically relies on feature point correspondence, the epipolar constraint can also be used for improving the accuracy of correspondence. We propose a probabilistic framework for estimating the epiploar geometry, in which the geometry and the feature correspondence are estimated iteratively at the same time. Using the EM algorithm to maximize a posteriori, our approach updates feature correspondence with estimated epipolar geometry. The correspondence is further improved with local diffusion on a prior Markov Random Field model. In turn, more accurate epipolar geometry is recovered. Experiments show this approach produces more accurate fundamental matrix compared with typical methods and can handle some challenging situations such as view rotation and scale changes.

AB - Finding epipolar geometry for two images is a fundamental problem in computer vision. While this typically relies on feature point correspondence, the epipolar constraint can also be used for improving the accuracy of correspondence. We propose a probabilistic framework for estimating the epiploar geometry, in which the geometry and the feature correspondence are estimated iteratively at the same time. Using the EM algorithm to maximize a posteriori, our approach updates feature correspondence with estimated epipolar geometry. The correspondence is further improved with local diffusion on a prior Markov Random Field model. In turn, more accurate epipolar geometry is recovered. Experiments show this approach produces more accurate fundamental matrix compared with typical methods and can handle some challenging situations such as view rotation and scale changes.

KW - EM algorithm

KW - Epipolar geometry

KW - Local diffusion

KW - MAP

UR - http://www.scopus.com/inward/record.url?scp=48149107375&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=48149107375&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2007.4379800

DO - 10.1109/ICIP.2007.4379800

M3 - Conference contribution

SN - 1424414377

SN - 9781424414376

VL - 5

BT - Proceedings - International Conference on Image Processing, ICIP

ER -