A maximum-likelihood surface estimator for dense range data

Ross T. Whitaker, Jens Gregor

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

This paper describes how to estimate 3D surface models from dense sets of noisy range data taken from different points of view, i.e., multiple range maps. The proposed method uses a sensor model to develop an expression for the likelihood of a 3D surface, conditional on a set of noisy range measurements. Optimizing this likelihood with respect to the model parameters provides an unbiased and efficient estimator. The proposed numerical algorithms make this estimation computationally practical for a wide variety of circumstances. The results from this method compare favorably with state-of-the-art approaches that rely on the closest-point or perpendicular distance metric, a convenient heuristic that produces biased solutions and fails completely when surfaces are not sufficiently smooth, as in the case of complex scenes or noisy range measurements. Empirical results on both simulated and real ladar data demonstrate the effectiveness of the proposed method for several different types of problems. Furthermore, the proposed method offers a general framework that can accommodate extensions to include surface priors (i.e., maximum a posteriori), more sophisticated noise models, and other sensing modalities, such as sonar or synthetic aperture radar.

Original languageEnglish (US)
Pages (from-to)1372-1387
Number of pages16
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume24
Issue number10
DOIs
StatePublished - Oct 1 2002

Fingerprint

Maximum likelihood
Maximum Likelihood
Estimator
Range of data
Likelihood
Ladar
Efficient Estimator
Sonar
Distance Metric
Maximum a Posteriori
Unbiased estimator
Synthetic Aperture
Optical radar
Synthetic aperture radar
Numerical Algorithms
Model
Perpendicular
Modality
Radar
Biased

All Science Journal Classification (ASJC) codes

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

Cite this

A maximum-likelihood surface estimator for dense range data. / Whitaker, Ross T.; Gregor, Jens.

In: IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 24, No. 10, 01.10.2002, p. 1372-1387.

Research output: Contribution to journalArticle

@article{65312cf4eca54333a1589cf1cffbb7ef,
title = "A maximum-likelihood surface estimator for dense range data",
abstract = "This paper describes how to estimate 3D surface models from dense sets of noisy range data taken from different points of view, i.e., multiple range maps. The proposed method uses a sensor model to develop an expression for the likelihood of a 3D surface, conditional on a set of noisy range measurements. Optimizing this likelihood with respect to the model parameters provides an unbiased and efficient estimator. The proposed numerical algorithms make this estimation computationally practical for a wide variety of circumstances. The results from this method compare favorably with state-of-the-art approaches that rely on the closest-point or perpendicular distance metric, a convenient heuristic that produces biased solutions and fails completely when surfaces are not sufficiently smooth, as in the case of complex scenes or noisy range measurements. Empirical results on both simulated and real ladar data demonstrate the effectiveness of the proposed method for several different types of problems. Furthermore, the proposed method offers a general framework that can accommodate extensions to include surface priors (i.e., maximum a posteriori), more sophisticated noise models, and other sensing modalities, such as sonar or synthetic aperture radar.",
author = "Whitaker, {Ross T.} and Jens Gregor",
year = "2002",
month = "10",
day = "1",
doi = "10.1109/TPAMI.2002.1039208",
language = "English (US)",
volume = "24",
pages = "1372--1387",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "10",

}

TY - JOUR

T1 - A maximum-likelihood surface estimator for dense range data

AU - Whitaker, Ross T.

AU - Gregor, Jens

PY - 2002/10/1

Y1 - 2002/10/1

N2 - This paper describes how to estimate 3D surface models from dense sets of noisy range data taken from different points of view, i.e., multiple range maps. The proposed method uses a sensor model to develop an expression for the likelihood of a 3D surface, conditional on a set of noisy range measurements. Optimizing this likelihood with respect to the model parameters provides an unbiased and efficient estimator. The proposed numerical algorithms make this estimation computationally practical for a wide variety of circumstances. The results from this method compare favorably with state-of-the-art approaches that rely on the closest-point or perpendicular distance metric, a convenient heuristic that produces biased solutions and fails completely when surfaces are not sufficiently smooth, as in the case of complex scenes or noisy range measurements. Empirical results on both simulated and real ladar data demonstrate the effectiveness of the proposed method for several different types of problems. Furthermore, the proposed method offers a general framework that can accommodate extensions to include surface priors (i.e., maximum a posteriori), more sophisticated noise models, and other sensing modalities, such as sonar or synthetic aperture radar.

AB - This paper describes how to estimate 3D surface models from dense sets of noisy range data taken from different points of view, i.e., multiple range maps. The proposed method uses a sensor model to develop an expression for the likelihood of a 3D surface, conditional on a set of noisy range measurements. Optimizing this likelihood with respect to the model parameters provides an unbiased and efficient estimator. The proposed numerical algorithms make this estimation computationally practical for a wide variety of circumstances. The results from this method compare favorably with state-of-the-art approaches that rely on the closest-point or perpendicular distance metric, a convenient heuristic that produces biased solutions and fails completely when surfaces are not sufficiently smooth, as in the case of complex scenes or noisy range measurements. Empirical results on both simulated and real ladar data demonstrate the effectiveness of the proposed method for several different types of problems. Furthermore, the proposed method offers a general framework that can accommodate extensions to include surface priors (i.e., maximum a posteriori), more sophisticated noise models, and other sensing modalities, such as sonar or synthetic aperture radar.

UR - http://www.scopus.com/inward/record.url?scp=0036807139&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036807139&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.2002.1039208

DO - 10.1109/TPAMI.2002.1039208

M3 - Article

VL - 24

SP - 1372

EP - 1387

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 10

ER -