Multiscale Representation of 3D Surfaces via Stochastic Mesh Laplacian

Ran Song, Liping Wang

Research output: Contribution to journalArticleResearchpeer-review

Abstract

Multiscale representation of a 3D surface mesh is a useful tool to understand a mesh both locally and globally. One method is to analyse eigenvalues and eigenvectors of some matrix which represents a discrete operator (e.g. the Laplacian)taking into account the topological and/or geometric structure of the input mesh. However, eigendecomposition is computationally expensive, making this method intractable for meshes containing more than a few thousand vertices. To overcome this problem, we present a novel method for multiscale mesh representation which avoids solving an eigenproblem, based on the proposed stochastic mesh Laplacian. We present the complete algorithm and the theoretical analysis of the stochastic mesh Laplacian. In the experiments, we compare our method with several state-of-the-art approaches to demonstrate its advantages over popular frameworks such as spectral mesh processing, heat diffusion and wavelets. The utility of the method is demonstrated via applications in mesh saliency and interest point detection.

Original languageEnglish
Pages (from-to)98-110
Number of pages13
JournalComputer-Aided Design
Volume115
DOIs
Publication statusPublished - 28 May 2019

Fingerprint

Eigenvalues and eigenfunctions
Processing
Experiments
Hot Temperature

Keywords

  • Laplacian
  • Mesh saliency
  • Multiscale representation
  • Stochastic matrix

Cite this

@article{c0994d0bebae44b590da1be28e5ccc4d,
title = "Multiscale Representation of 3D Surfaces via Stochastic Mesh Laplacian",
abstract = "Multiscale representation of a 3D surface mesh is a useful tool to understand a mesh both locally and globally. One method is to analyse eigenvalues and eigenvectors of some matrix which represents a discrete operator (e.g. the Laplacian)taking into account the topological and/or geometric structure of the input mesh. However, eigendecomposition is computationally expensive, making this method intractable for meshes containing more than a few thousand vertices. To overcome this problem, we present a novel method for multiscale mesh representation which avoids solving an eigenproblem, based on the proposed stochastic mesh Laplacian. We present the complete algorithm and the theoretical analysis of the stochastic mesh Laplacian. In the experiments, we compare our method with several state-of-the-art approaches to demonstrate its advantages over popular frameworks such as spectral mesh processing, heat diffusion and wavelets. The utility of the method is demonstrated via applications in mesh saliency and interest point detection.",
keywords = "Laplacian, Mesh saliency, Multiscale representation, Stochastic matrix",
author = "Ran Song and Liping Wang",
year = "2019",
month = "5",
day = "28",
doi = "10.1016/j.cad.2019.05.006",
language = "English",
volume = "115",
pages = "98--110",
journal = "Computer-Aided Design",
issn = "0010-4485",
publisher = "Elsevier",

}

Multiscale Representation of 3D Surfaces via Stochastic Mesh Laplacian. / Song, Ran; Wang, Liping.

In: Computer-Aided Design, Vol. 115, 28.05.2019, p. 98-110.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Multiscale Representation of 3D Surfaces via Stochastic Mesh Laplacian

AU - Song, Ran

AU - Wang, Liping

PY - 2019/5/28

Y1 - 2019/5/28

N2 - Multiscale representation of a 3D surface mesh is a useful tool to understand a mesh both locally and globally. One method is to analyse eigenvalues and eigenvectors of some matrix which represents a discrete operator (e.g. the Laplacian)taking into account the topological and/or geometric structure of the input mesh. However, eigendecomposition is computationally expensive, making this method intractable for meshes containing more than a few thousand vertices. To overcome this problem, we present a novel method for multiscale mesh representation which avoids solving an eigenproblem, based on the proposed stochastic mesh Laplacian. We present the complete algorithm and the theoretical analysis of the stochastic mesh Laplacian. In the experiments, we compare our method with several state-of-the-art approaches to demonstrate its advantages over popular frameworks such as spectral mesh processing, heat diffusion and wavelets. The utility of the method is demonstrated via applications in mesh saliency and interest point detection.

AB - Multiscale representation of a 3D surface mesh is a useful tool to understand a mesh both locally and globally. One method is to analyse eigenvalues and eigenvectors of some matrix which represents a discrete operator (e.g. the Laplacian)taking into account the topological and/or geometric structure of the input mesh. However, eigendecomposition is computationally expensive, making this method intractable for meshes containing more than a few thousand vertices. To overcome this problem, we present a novel method for multiscale mesh representation which avoids solving an eigenproblem, based on the proposed stochastic mesh Laplacian. We present the complete algorithm and the theoretical analysis of the stochastic mesh Laplacian. In the experiments, we compare our method with several state-of-the-art approaches to demonstrate its advantages over popular frameworks such as spectral mesh processing, heat diffusion and wavelets. The utility of the method is demonstrated via applications in mesh saliency and interest point detection.

KW - Laplacian

KW - Mesh saliency

KW - Multiscale representation

KW - Stochastic matrix

UR - http://www.scopus.com/inward/record.url?scp=85066235231&partnerID=8YFLogxK

U2 - 10.1016/j.cad.2019.05.006

DO - 10.1016/j.cad.2019.05.006

M3 - Article

VL - 115

SP - 98

EP - 110

JO - Computer-Aided Design

JF - Computer-Aided Design

SN - 0010-4485

ER -