It is well known that both shape and motion can be factorized directly from the measurement matrix constructed from feature points trajectories under orthographic camera model. In practical applications, the measurement matrix might be contaminated by noises and contains outliers. A direct SVD (Singular Value Decomposition) to the measurement matrix with outliers would yield erroneous result. This paper presents a novel algorithm for computing SVD with outliers. We decompose the SVD computation as a set of alternate linear regression subproblems. The linear regression subproblems are solved robustly by applying the RANSAC strategy. The proposed robust factorization method with outliers can improve the reconstruction result remarkably. Quantitative and qualitative experiments illustrate the good performance of the proposed method.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Xi LI, Zhengnan NING, Liuwei XIANG, "Robust 3D Reconstruction with Outliers Using RANSAC Based Singular Value Decomposition" in IEICE TRANSACTIONS on Information,
vol. E88-D, no. 8, pp. 2001-2004, August 2005, doi: 10.1093/ietisy/e88-d.8.2001.
Abstract: It is well known that both shape and motion can be factorized directly from the measurement matrix constructed from feature points trajectories under orthographic camera model. In practical applications, the measurement matrix might be contaminated by noises and contains outliers. A direct SVD (Singular Value Decomposition) to the measurement matrix with outliers would yield erroneous result. This paper presents a novel algorithm for computing SVD with outliers. We decompose the SVD computation as a set of alternate linear regression subproblems. The linear regression subproblems are solved robustly by applying the RANSAC strategy. The proposed robust factorization method with outliers can improve the reconstruction result remarkably. Quantitative and qualitative experiments illustrate the good performance of the proposed method.
URL: https://global.ieice.org/en_transactions/information/10.1093/ietisy/e88-d.8.2001/_p
Copy
@ARTICLE{e88-d_8_2001,
author={Xi LI, Zhengnan NING, Liuwei XIANG, },
journal={IEICE TRANSACTIONS on Information},
title={Robust 3D Reconstruction with Outliers Using RANSAC Based Singular Value Decomposition},
year={2005},
volume={E88-D},
number={8},
pages={2001-2004},
abstract={It is well known that both shape and motion can be factorized directly from the measurement matrix constructed from feature points trajectories under orthographic camera model. In practical applications, the measurement matrix might be contaminated by noises and contains outliers. A direct SVD (Singular Value Decomposition) to the measurement matrix with outliers would yield erroneous result. This paper presents a novel algorithm for computing SVD with outliers. We decompose the SVD computation as a set of alternate linear regression subproblems. The linear regression subproblems are solved robustly by applying the RANSAC strategy. The proposed robust factorization method with outliers can improve the reconstruction result remarkably. Quantitative and qualitative experiments illustrate the good performance of the proposed method.},
keywords={},
doi={10.1093/ietisy/e88-d.8.2001},
ISSN={},
month={August},}
Copy
TY - JOUR
TI - Robust 3D Reconstruction with Outliers Using RANSAC Based Singular Value Decomposition
T2 - IEICE TRANSACTIONS on Information
SP - 2001
EP - 2004
AU - Xi LI
AU - Zhengnan NING
AU - Liuwei XIANG
PY - 2005
DO - 10.1093/ietisy/e88-d.8.2001
JO - IEICE TRANSACTIONS on Information
SN -
VL - E88-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2005
AB - It is well known that both shape and motion can be factorized directly from the measurement matrix constructed from feature points trajectories under orthographic camera model. In practical applications, the measurement matrix might be contaminated by noises and contains outliers. A direct SVD (Singular Value Decomposition) to the measurement matrix with outliers would yield erroneous result. This paper presents a novel algorithm for computing SVD with outliers. We decompose the SVD computation as a set of alternate linear regression subproblems. The linear regression subproblems are solved robustly by applying the RANSAC strategy. The proposed robust factorization method with outliers can improve the reconstruction result remarkably. Quantitative and qualitative experiments illustrate the good performance of the proposed method.
ER -