3D laplacian pyramid signature

Kaimo Hu, Yi Fang

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

Abstract

We introduce a simple and effective point descriptor, called 3D Laplacian Pyramid Signature (3DLPS), by extending and adapting the Laplacian Pyramid defined in 2D images to 3D shapes. The signature is represented as a high-dimensional feature vector recording the magnitudes of mean curvatures, which are captured through sequentially applying Laplacian of Gaussian (LOG) operators on each vertex of 3D shapes. We show that 3DLPS organizes the intrinsic geometry information concisely, while possessing high sensitivity and specificity. Compared with existing point signatures, 3DLPS is robust and easy to compute, yet captures enough information embedded in the shape. We describe how 3DLPS may potentially benefit the applications involved in shape analysis, and especially demonstrate how to incorporate it in point correspondence detection, best view selection and automatic mesh segmentation. Experiments across a collection of shapes have verified its effectiveness.

Original languageEnglish (US)
Title of host publicationComputer Vision - 12th Asian Conference on Computer Vision, ACCV 2014, Revised Selected Papers
PublisherSpringer-Verlag
Pages306-321
Number of pages16
ISBN (Print)9783319166339
DOIs
StatePublished - Jan 1 2015
Event12th Asian Conference on Computer Vision, ACCV 2014 - Singapore, Singapore
Duration: Nov 1 2014Nov 5 2014

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9010
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other12th Asian Conference on Computer Vision, ACCV 2014
CountrySingapore
CitySingapore
Period11/1/1411/5/14

Fingerprint

Pyramid
Signature
Geometry
3D shape
Experiments
Information Geometry
Shape Analysis
Mean Curvature
Feature Vector
Descriptors
Specificity
High-dimensional
Correspondence
Segmentation
Mesh
Vertex of a graph
Operator
Demonstrate
Experiment

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Hu, K., & Fang, Y. (2015). 3D laplacian pyramid signature. In Computer Vision - 12th Asian Conference on Computer Vision, ACCV 2014, Revised Selected Papers (pp. 306-321). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9010). Springer-Verlag. https://doi.org/10.1007/978-3-319-16634-6_23

3D laplacian pyramid signature. / Hu, Kaimo; Fang, Yi.

Computer Vision - 12th Asian Conference on Computer Vision, ACCV 2014, Revised Selected Papers. Springer-Verlag, 2015. p. 306-321 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9010).

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

Hu, K & Fang, Y 2015, 3D laplacian pyramid signature. in Computer Vision - 12th Asian Conference on Computer Vision, ACCV 2014, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9010, Springer-Verlag, pp. 306-321, 12th Asian Conference on Computer Vision, ACCV 2014, Singapore, Singapore, 11/1/14. https://doi.org/10.1007/978-3-319-16634-6_23
Hu K, Fang Y. 3D laplacian pyramid signature. In Computer Vision - 12th Asian Conference on Computer Vision, ACCV 2014, Revised Selected Papers. Springer-Verlag. 2015. p. 306-321. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-16634-6_23
Hu, Kaimo ; Fang, Yi. / 3D laplacian pyramid signature. Computer Vision - 12th Asian Conference on Computer Vision, ACCV 2014, Revised Selected Papers. Springer-Verlag, 2015. pp. 306-321 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{2f953b51249443b69786961d216613fe,
title = "3D laplacian pyramid signature",
abstract = "We introduce a simple and effective point descriptor, called 3D Laplacian Pyramid Signature (3DLPS), by extending and adapting the Laplacian Pyramid defined in 2D images to 3D shapes. The signature is represented as a high-dimensional feature vector recording the magnitudes of mean curvatures, which are captured through sequentially applying Laplacian of Gaussian (LOG) operators on each vertex of 3D shapes. We show that 3DLPS organizes the intrinsic geometry information concisely, while possessing high sensitivity and specificity. Compared with existing point signatures, 3DLPS is robust and easy to compute, yet captures enough information embedded in the shape. We describe how 3DLPS may potentially benefit the applications involved in shape analysis, and especially demonstrate how to incorporate it in point correspondence detection, best view selection and automatic mesh segmentation. Experiments across a collection of shapes have verified its effectiveness.",
author = "Kaimo Hu and Yi Fang",
year = "2015",
month = "1",
day = "1",
doi = "10.1007/978-3-319-16634-6_23",
language = "English (US)",
isbn = "9783319166339",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer-Verlag",
pages = "306--321",
booktitle = "Computer Vision - 12th Asian Conference on Computer Vision, ACCV 2014, Revised Selected Papers",

}

TY - GEN

T1 - 3D laplacian pyramid signature

AU - Hu, Kaimo

AU - Fang, Yi

PY - 2015/1/1

Y1 - 2015/1/1

N2 - We introduce a simple and effective point descriptor, called 3D Laplacian Pyramid Signature (3DLPS), by extending and adapting the Laplacian Pyramid defined in 2D images to 3D shapes. The signature is represented as a high-dimensional feature vector recording the magnitudes of mean curvatures, which are captured through sequentially applying Laplacian of Gaussian (LOG) operators on each vertex of 3D shapes. We show that 3DLPS organizes the intrinsic geometry information concisely, while possessing high sensitivity and specificity. Compared with existing point signatures, 3DLPS is robust and easy to compute, yet captures enough information embedded in the shape. We describe how 3DLPS may potentially benefit the applications involved in shape analysis, and especially demonstrate how to incorporate it in point correspondence detection, best view selection and automatic mesh segmentation. Experiments across a collection of shapes have verified its effectiveness.

AB - We introduce a simple and effective point descriptor, called 3D Laplacian Pyramid Signature (3DLPS), by extending and adapting the Laplacian Pyramid defined in 2D images to 3D shapes. The signature is represented as a high-dimensional feature vector recording the magnitudes of mean curvatures, which are captured through sequentially applying Laplacian of Gaussian (LOG) operators on each vertex of 3D shapes. We show that 3DLPS organizes the intrinsic geometry information concisely, while possessing high sensitivity and specificity. Compared with existing point signatures, 3DLPS is robust and easy to compute, yet captures enough information embedded in the shape. We describe how 3DLPS may potentially benefit the applications involved in shape analysis, and especially demonstrate how to incorporate it in point correspondence detection, best view selection and automatic mesh segmentation. Experiments across a collection of shapes have verified its effectiveness.

UR - http://www.scopus.com/inward/record.url?scp=84942515649&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84942515649&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-16634-6_23

DO - 10.1007/978-3-319-16634-6_23

M3 - Conference contribution

SN - 9783319166339

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 306

EP - 321

BT - Computer Vision - 12th Asian Conference on Computer Vision, ACCV 2014, Revised Selected Papers

PB - Springer-Verlag

ER -