Robust Scale Estimation from Ensemble Inlier Sets for Random Sample Consensus Methods
Abstract
This paper proposes a RANSAC modification that performs automatic estimation of the scale of inlier noise. The scale estimation takes advantage of accumulated inlier sets from all proposed models. It is shown that the proposed method gives robust results in case of high outlier ratio data, in spite that no user specified threshold is needed. The method also improves sampling efficiency, without requiring any auxiliary information other than the data to be modeled.
Cite
Text
Fan and Pylvänäinen. "Robust Scale Estimation from Ensemble Inlier Sets for Random Sample Consensus Methods." European Conference on Computer Vision, 2008. doi:10.1007/978-3-540-88690-7_14Markdown
[Fan and Pylvänäinen. "Robust Scale Estimation from Ensemble Inlier Sets for Random Sample Consensus Methods." European Conference on Computer Vision, 2008.](https://mlanthology.org/eccv/2008/fan2008eccv-robust/) doi:10.1007/978-3-540-88690-7_14BibTeX
@inproceedings{fan2008eccv-robust,
title = {{Robust Scale Estimation from Ensemble Inlier Sets for Random Sample Consensus Methods}},
author = {Fan, Lixin and Pylvänäinen, Timo},
booktitle = {European Conference on Computer Vision},
year = {2008},
pages = {182-195},
doi = {10.1007/978-3-540-88690-7_14},
url = {https://mlanthology.org/eccv/2008/fan2008eccv-robust/}
}