A generalized kernel consensus-based robust estimator

Hanzi Wang; Daniel Mirota; Gregory D. Hager

doi:10.1109/TPAMI.2009.148

A generalized kernel consensus-based robust estimator

Hanzi Wang, Daniel Mirota, Gregory D. Hager

Whiting School of Engineering

Research output: Contribution to journal › Article › peer-review

46 Scopus citations

Abstract

In this paper, we present a new Adaptive-Scale Kernel Consensus (ASKC) robust estimator as a generalization of the popular and state-of-the-art robust estimators such as RANdom SAmple Consensus (RANSAC), Adaptive Scale Sample Consensus (ASSC), and Maximum Kernel Density Estimator (MKDE). The ASKC framework is grounded on and unifies these robust estimators using nonparametric kernel density estimation theory. In particular, we show that each of these methods is a special case of ASKC using a specific kernel. Like these methods, ASKC can tolerate more than 50 percent outliers, but it can also automatically estimate the scale of inliers. We apply ASKC to two important areas in computer vision, robust motion estimation and pose estimation, and show comparative results on both synthetic and real data.

Original language	English (US)
Article number	5184846
Pages (from-to)	178-184
Number of pages	7
Journal	IEEE Transactions on Pattern Analysis and Machine Intelligence
Volume	32
Issue number	1
DOIs	https://doi.org/10.1109/TPAMI.2009.148
State	Published - 2010

Keywords

Kernel density estimation
Model fitting
Motion estimation
Pose estimation
Robust statistics

ASJC Scopus subject areas

Software
Computer Vision and Pattern Recognition
Computational Theory and Mathematics
Artificial Intelligence
Applied Mathematics

Access to Document

10.1109/TPAMI.2009.148

Cite this

@article{078c572a238d44989f15744f2ac16f6c,

title = "A generalized kernel consensus-based robust estimator",

abstract = "In this paper, we present a new Adaptive-Scale Kernel Consensus (ASKC) robust estimator as a generalization of the popular and state-of-the-art robust estimators such as RANdom SAmple Consensus (RANSAC), Adaptive Scale Sample Consensus (ASSC), and Maximum Kernel Density Estimator (MKDE). The ASKC framework is grounded on and unifies these robust estimators using nonparametric kernel density estimation theory. In particular, we show that each of these methods is a special case of ASKC using a specific kernel. Like these methods, ASKC can tolerate more than 50 percent outliers, but it can also automatically estimate the scale of inliers. We apply ASKC to two important areas in computer vision, robust motion estimation and pose estimation, and show comparative results on both synthetic and real data.",

keywords = "Kernel density estimation, Model fitting, Motion estimation, Pose estimation, Robust statistics",

author = "Hanzi Wang and Daniel Mirota and Hager, {Gregory D.}",

note = "Funding Information: Parts of the work described in this paper appeared in [26], [27]. The authors thank the anonymous reviewers of those papers, and the TPAMI reviewers, Professors David Suter and Russell H. Taylor for their valuable comments and suggestions. The authors also thank Professor Darius Burschka for providing the endoscopic sinus image data. The work was supported by the US National",

year = "2010",

doi = "10.1109/TPAMI.2009.148",

language = "English (US)",

volume = "32",

pages = "178--184",

journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",

issn = "0162-8828",

publisher = "IEEE Computer Society",

number = "1",

}

TY - JOUR

T1 - A generalized kernel consensus-based robust estimator

AU - Wang, Hanzi

AU - Mirota, Daniel

AU - Hager, Gregory D.

N1 - Funding Information: Parts of the work described in this paper appeared in [26], [27]. The authors thank the anonymous reviewers of those papers, and the TPAMI reviewers, Professors David Suter and Russell H. Taylor for their valuable comments and suggestions. The authors also thank Professor Darius Burschka for providing the endoscopic sinus image data. The work was supported by the US National

PY - 2010

Y1 - 2010

N2 - In this paper, we present a new Adaptive-Scale Kernel Consensus (ASKC) robust estimator as a generalization of the popular and state-of-the-art robust estimators such as RANdom SAmple Consensus (RANSAC), Adaptive Scale Sample Consensus (ASSC), and Maximum Kernel Density Estimator (MKDE). The ASKC framework is grounded on and unifies these robust estimators using nonparametric kernel density estimation theory. In particular, we show that each of these methods is a special case of ASKC using a specific kernel. Like these methods, ASKC can tolerate more than 50 percent outliers, but it can also automatically estimate the scale of inliers. We apply ASKC to two important areas in computer vision, robust motion estimation and pose estimation, and show comparative results on both synthetic and real data.

AB - In this paper, we present a new Adaptive-Scale Kernel Consensus (ASKC) robust estimator as a generalization of the popular and state-of-the-art robust estimators such as RANdom SAmple Consensus (RANSAC), Adaptive Scale Sample Consensus (ASSC), and Maximum Kernel Density Estimator (MKDE). The ASKC framework is grounded on and unifies these robust estimators using nonparametric kernel density estimation theory. In particular, we show that each of these methods is a special case of ASKC using a specific kernel. Like these methods, ASKC can tolerate more than 50 percent outliers, but it can also automatically estimate the scale of inliers. We apply ASKC to two important areas in computer vision, robust motion estimation and pose estimation, and show comparative results on both synthetic and real data.

KW - Kernel density estimation

KW - Model fitting

KW - Motion estimation

KW - Pose estimation

KW - Robust statistics

UR - http://www.scopus.com/inward/record.url?scp=74949108100&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=74949108100&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.2009.148

DO - 10.1109/TPAMI.2009.148

M3 - Article

C2 - 19926908

AN - SCOPUS:74949108100

SN - 0162-8828

VL - 32

SP - 178

EP - 184

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

IS - 1

M1 - 5184846

ER -

A generalized kernel consensus-based robust estimator

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this