noRANSAC for fundamental matrix estimation

Risultato della ricerca: Conference contribution

5 Citazioni (Scopus)

Abstract

The estimation of the fundamental matrix from a set of corresponding points is a relevant topic in epipolar stereo geometry [10]. Due to the high amount of outliers between the matches, RANSAC-based approaches [7, 13, 29] have been used to obtain the fundamental matrix. In this paper two new contributes are presented: a new normalized epipolar error measure which takes into account the shape of the features used as matches [17] and a new strategy to compare fundamental matrices. The proposed error measure gives good results and it does not depend on the image scale. Moreover, the new evaluation strategy describes a valid tool to compare different RANSAC-based methods because it does not rely on the inlier ratio, which could not correspond to the best allowable fundamental matrix estimated model, but it makes use of a reference ground truth fundamental matrix obtained by a set of corresponding points given by the user
Lingua originaleEnglish
Titolo della pubblicazione ospiteProceedings of 22th British Machine Vision Conference, BMVC 2011
Pagine1-11
Numero di pagine11
Stato di pubblicazionePublished - 2011

Fingerprint

Geometry

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Cita questo

Tegolo, D., & Bellavia, F. (2011). noRANSAC for fundamental matrix estimation. In Proceedings of 22th British Machine Vision Conference, BMVC 2011 (pagg. 1-11)

noRANSAC for fundamental matrix estimation. / Tegolo, Domenico; Bellavia, Fabio.

Proceedings of 22th British Machine Vision Conference, BMVC 2011. 2011. pag. 1-11.

Risultato della ricerca: Conference contribution

Tegolo, D & Bellavia, F 2011, noRANSAC for fundamental matrix estimation. in Proceedings of 22th British Machine Vision Conference, BMVC 2011. pagg. 1-11.
Tegolo D, Bellavia F. noRANSAC for fundamental matrix estimation. In Proceedings of 22th British Machine Vision Conference, BMVC 2011. 2011. pag. 1-11
Tegolo, Domenico ; Bellavia, Fabio. / noRANSAC for fundamental matrix estimation. Proceedings of 22th British Machine Vision Conference, BMVC 2011. 2011. pagg. 1-11
@inproceedings{0885ca65ecb841b48e601d395ba77ff8,
title = "noRANSAC for fundamental matrix estimation",
abstract = "The estimation of the fundamental matrix from a set of corresponding points is a relevant topic in epipolar stereo geometry [10]. Due to the high amount of outliers between the matches, RANSAC-based approaches [7, 13, 29] have been used to obtain the fundamental matrix. In this paper two new contributes are presented: a new normalized epipolar error measure which takes into account the shape of the features used as matches [17] and a new strategy to compare fundamental matrices. The proposed error measure gives good results and it does not depend on the image scale. Moreover, the new evaluation strategy describes a valid tool to compare different RANSAC-based methods because it does not rely on the inlier ratio, which could not correspond to the best allowable fundamental matrix estimated model, but it makes use of a reference ground truth fundamental matrix obtained by a set of corresponding points given by the user",
author = "Domenico Tegolo and Fabio Bellavia",
year = "2011",
language = "English",
isbn = "1-901725-43-X",
pages = "1--11",
booktitle = "Proceedings of 22th British Machine Vision Conference, BMVC 2011",

}

TY - GEN

T1 - noRANSAC for fundamental matrix estimation

AU - Tegolo, Domenico

AU - Bellavia, Fabio

PY - 2011

Y1 - 2011

N2 - The estimation of the fundamental matrix from a set of corresponding points is a relevant topic in epipolar stereo geometry [10]. Due to the high amount of outliers between the matches, RANSAC-based approaches [7, 13, 29] have been used to obtain the fundamental matrix. In this paper two new contributes are presented: a new normalized epipolar error measure which takes into account the shape of the features used as matches [17] and a new strategy to compare fundamental matrices. The proposed error measure gives good results and it does not depend on the image scale. Moreover, the new evaluation strategy describes a valid tool to compare different RANSAC-based methods because it does not rely on the inlier ratio, which could not correspond to the best allowable fundamental matrix estimated model, but it makes use of a reference ground truth fundamental matrix obtained by a set of corresponding points given by the user

AB - The estimation of the fundamental matrix from a set of corresponding points is a relevant topic in epipolar stereo geometry [10]. Due to the high amount of outliers between the matches, RANSAC-based approaches [7, 13, 29] have been used to obtain the fundamental matrix. In this paper two new contributes are presented: a new normalized epipolar error measure which takes into account the shape of the features used as matches [17] and a new strategy to compare fundamental matrices. The proposed error measure gives good results and it does not depend on the image scale. Moreover, the new evaluation strategy describes a valid tool to compare different RANSAC-based methods because it does not rely on the inlier ratio, which could not correspond to the best allowable fundamental matrix estimated model, but it makes use of a reference ground truth fundamental matrix obtained by a set of corresponding points given by the user

UR - http://hdl.handle.net/10447/60416

UR - http://www.bmva.org/bmvc/2011/proceedings/paper98/index.html

M3 - Conference contribution

SN - 1-901725-43-X

SP - 1

EP - 11

BT - Proceedings of 22th British Machine Vision Conference, BMVC 2011

ER -