Fast, robust, and consistent camera motion estimation

Tong Zhang; Carlo Tomasi

doi:10.1109/CVPR.1999.786934

Fast, robust, and consistent camera motion estimation

Tong Zhang^*, Carlo Tomasi

^*Corresponding author for this work

Research output: Contribution to journal › Conference article published in journal › peer-review

37 Citations (Scopus)

Abstract

Previous algorithms that recover camera motion from image velocities suffer from both bias and excessive variance in the results. We propose a robust estimator of camera motion that is statistically consistent when image noise is isotropic. Consistency means that the estimated motion converges in probability to the true value as the number of image points increases. An algorithm based on reweighted Gauss-Newton iterations handles 100 velocity measurements in about 50 milliseconds on a workstation.

Original language	English
Pages (from-to)	164-170
Number of pages	7
Journal	Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Volume	1
DOIs	https://doi.org/10.1109/CVPR.1999.786934
Publication status	Published - 1999
Externally published	Yes
Event	Proceedings of the 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'99) - Fort Collins, CO, USA Duration: 23 Jun 1999 → 25 Jun 1999

Access to Document

10.1109/CVPR.1999.786934

Cite this

@article{8c79cd436d284afebd480218025e984f,

title = "Fast, robust, and consistent camera motion estimation",

abstract = "Previous algorithms that recover camera motion from image velocities suffer from both bias and excessive variance in the results. We propose a robust estimator of camera motion that is statistically consistent when image noise is isotropic. Consistency means that the estimated motion converges in probability to the true value as the number of image points increases. An algorithm based on reweighted Gauss-Newton iterations handles 100 velocity measurements in about 50 milliseconds on a workstation.",

author = "Tong Zhang and Carlo Tomasi",

year = "1999",

doi = "10.1109/CVPR.1999.786934",

language = "English",

volume = "1",

pages = "164--170",

journal = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",

issn = "1063-6919",

publisher = "IEEE Computer Society",

note = "Proceedings of the 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'99) ; Conference date: 23-06-1999 Through 25-06-1999",

}

TY - JOUR

T1 - Fast, robust, and consistent camera motion estimation

AU - Zhang, Tong

AU - Tomasi, Carlo

PY - 1999

Y1 - 1999

N2 - Previous algorithms that recover camera motion from image velocities suffer from both bias and excessive variance in the results. We propose a robust estimator of camera motion that is statistically consistent when image noise is isotropic. Consistency means that the estimated motion converges in probability to the true value as the number of image points increases. An algorithm based on reweighted Gauss-Newton iterations handles 100 velocity measurements in about 50 milliseconds on a workstation.

AB - Previous algorithms that recover camera motion from image velocities suffer from both bias and excessive variance in the results. We propose a robust estimator of camera motion that is statistically consistent when image noise is isotropic. Consistency means that the estimated motion converges in probability to the true value as the number of image points increases. An algorithm based on reweighted Gauss-Newton iterations handles 100 velocity measurements in about 50 milliseconds on a workstation.

UR - https://www.scopus.com/pages/publications/0032667534

U2 - 10.1109/CVPR.1999.786934

DO - 10.1109/CVPR.1999.786934

M3 - Conference article published in journal

AN - SCOPUS:0032667534

SN - 1063-6919

VL - 1

SP - 164

EP - 170

JO - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

JF - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

T2 - Proceedings of the 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'99)

Y2 - 23 June 1999 through 25 June 1999

ER -

Fast, robust, and consistent camera motion estimation

Abstract

Access to Document

Other files and links

Fingerprint

Cite this