Optimal bandwidth selection for MLS surfaces

Hao Wang, Carlos E. Scheidegger, Cĺaudio T. Silva

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We address the problem of bandwidth selection in MLS surfaces. While the problem has received relatively little attention in the literature, we show that appropriate selection plays a critical role in the quality of reconstructed surfaces. We formulate the MLS polynomial fitting step as a kernel regression problem for both noiseless and noisy data. Based on this framework, we develop fast algorithms to find optimal bandwidths for a large class of weight functions. We show experimental comparisons of our method, which outperforms heuristically chosen functions and weights previously proposed. We conclude with a discussion of the implications of the Levin's two-step MLS projection for bandwidth selection.

Original languageEnglish (US)
Title of host publicationIEEE International Conference on Shape Modeling and Applications 2008, Proceedings, SMI
Pages111-120
Number of pages10
DOIs
StatePublished - 2008
EventIEEE International Conference on Shape Modeling and Applications 2008, SMI - Stony Brook, NY, United States
Duration: Jun 4 2008Jun 6 2008

Other

OtherIEEE International Conference on Shape Modeling and Applications 2008, SMI
CountryUnited States
CityStony Brook, NY
Period6/4/086/6/08

Fingerprint

Bandwidth
Polynomials

Keywords

  • Bandwidth
  • Kernel regression
  • MLS
  • Point cloud

ASJC Scopus subject areas

  • Computer Science Applications
  • Computational Theory and Mathematics

Cite this

Wang, H., Scheidegger, C. E., & Silva, C. T. (2008). Optimal bandwidth selection for MLS surfaces. In IEEE International Conference on Shape Modeling and Applications 2008, Proceedings, SMI (pp. 111-120). [4547957] https://doi.org/10.1109/SMI.2008.4547957

Optimal bandwidth selection for MLS surfaces. / Wang, Hao; Scheidegger, Carlos E.; Silva, Cĺaudio T.

IEEE International Conference on Shape Modeling and Applications 2008, Proceedings, SMI. 2008. p. 111-120 4547957.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wang, H, Scheidegger, CE & Silva, CT 2008, Optimal bandwidth selection for MLS surfaces. in IEEE International Conference on Shape Modeling and Applications 2008, Proceedings, SMI., 4547957, pp. 111-120, IEEE International Conference on Shape Modeling and Applications 2008, SMI, Stony Brook, NY, United States, 6/4/08. https://doi.org/10.1109/SMI.2008.4547957
Wang H, Scheidegger CE, Silva CT. Optimal bandwidth selection for MLS surfaces. In IEEE International Conference on Shape Modeling and Applications 2008, Proceedings, SMI. 2008. p. 111-120. 4547957 https://doi.org/10.1109/SMI.2008.4547957
Wang, Hao ; Scheidegger, Carlos E. ; Silva, Cĺaudio T. / Optimal bandwidth selection for MLS surfaces. IEEE International Conference on Shape Modeling and Applications 2008, Proceedings, SMI. 2008. pp. 111-120
@inproceedings{deee8bdaec0242e194a7801daca6a85d,
title = "Optimal bandwidth selection for MLS surfaces",
abstract = "We address the problem of bandwidth selection in MLS surfaces. While the problem has received relatively little attention in the literature, we show that appropriate selection plays a critical role in the quality of reconstructed surfaces. We formulate the MLS polynomial fitting step as a kernel regression problem for both noiseless and noisy data. Based on this framework, we develop fast algorithms to find optimal bandwidths for a large class of weight functions. We show experimental comparisons of our method, which outperforms heuristically chosen functions and weights previously proposed. We conclude with a discussion of the implications of the Levin's two-step MLS projection for bandwidth selection.",
keywords = "Bandwidth, Kernel regression, MLS, Point cloud",
author = "Hao Wang and Scheidegger, {Carlos E.} and Silva, {Cĺaudio T.}",
year = "2008",
doi = "10.1109/SMI.2008.4547957",
language = "English (US)",
isbn = "9781424422609",
pages = "111--120",
booktitle = "IEEE International Conference on Shape Modeling and Applications 2008, Proceedings, SMI",

}

TY - GEN

T1 - Optimal bandwidth selection for MLS surfaces

AU - Wang, Hao

AU - Scheidegger, Carlos E.

AU - Silva, Cĺaudio T.

PY - 2008

Y1 - 2008

N2 - We address the problem of bandwidth selection in MLS surfaces. While the problem has received relatively little attention in the literature, we show that appropriate selection plays a critical role in the quality of reconstructed surfaces. We formulate the MLS polynomial fitting step as a kernel regression problem for both noiseless and noisy data. Based on this framework, we develop fast algorithms to find optimal bandwidths for a large class of weight functions. We show experimental comparisons of our method, which outperforms heuristically chosen functions and weights previously proposed. We conclude with a discussion of the implications of the Levin's two-step MLS projection for bandwidth selection.

AB - We address the problem of bandwidth selection in MLS surfaces. While the problem has received relatively little attention in the literature, we show that appropriate selection plays a critical role in the quality of reconstructed surfaces. We formulate the MLS polynomial fitting step as a kernel regression problem for both noiseless and noisy data. Based on this framework, we develop fast algorithms to find optimal bandwidths for a large class of weight functions. We show experimental comparisons of our method, which outperforms heuristically chosen functions and weights previously proposed. We conclude with a discussion of the implications of the Levin's two-step MLS projection for bandwidth selection.

KW - Bandwidth

KW - Kernel regression

KW - MLS

KW - Point cloud

UR - http://www.scopus.com/inward/record.url?scp=50949118600&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=50949118600&partnerID=8YFLogxK

U2 - 10.1109/SMI.2008.4547957

DO - 10.1109/SMI.2008.4547957

M3 - Conference contribution

SN - 9781424422609

SP - 111

EP - 120

BT - IEEE International Conference on Shape Modeling and Applications 2008, Proceedings, SMI

ER -