Persistence-based Interest Point Detection for 3D Deformable Surface

Xupeng Wang, Ferdous Sohel, Mohammed Bennamoun, Yulan Guo, Hang Lei

Research output: Chapter in Book/Conference paperConference paper

101 Downloads (Pure)

Abstract

Several approaches for interest point detection on rigid shapes have been proposed, but few are available for non-rigid shapes. It is a very challenging task due to the presence of the large degrees of local deformations. This paper presents a novel method called persistence-based heat kernel signature (pHKS). It consists of two steps: scalar field construction and interest point detection. We propose to use the heat kernel signature function at a moderately small time scale to construct the scalar field. It has the advantage of being stable under various transformations. Based on the predefined scalar field, a 0-dimensional persistence diagram is computed, and the local geometric and global structural information of the shape are captured at the same time. Points with local maxima and high persistence are selected as interest points. We perform a comprehensive evaluation on two popular datasets (i.e., PHOTOMESH and Interest Points Dataset) to show the effectiveness of our method. Compared with existing techniques, our interest point detector achieves a superior performance in terms of repeatability and distinctiveness.
Original languageEnglish
Title of host publicationProceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
EditorsAna Paula Claudio, Dominique Bechmann, Jose Braz
Place of PublicationPortugal
PublisherScitepress
Pages58-69
Volume1
ISBN (Print)9789897582240
Publication statusPublished - 2017
Event 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Porto, Portugal
Duration: 27 Feb 20171 Mar 2017
http://www.grapp.visigrapp.org/?y=2017

Conference

Conference 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
CountryPortugal
CityPorto
Period27/02/171/03/17
Internet address

Fingerprint

Detectors
Hot Temperature

Cite this

Wang, X., Sohel, F., Bennamoun, M., Guo, Y., & Lei, H. (2017). Persistence-based Interest Point Detection for 3D Deformable Surface. In A. P. Claudio, D. Bechmann, & J. Braz (Eds.), Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (Vol. 1, pp. 58-69). Portugal: Scitepress.
Wang, Xupeng ; Sohel, Ferdous ; Bennamoun, Mohammed ; Guo, Yulan ; Lei, Hang. / Persistence-based Interest Point Detection for 3D Deformable Surface. Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications . editor / Ana Paula Claudio ; Dominique Bechmann ; Jose Braz. Vol. 1 Portugal : Scitepress, 2017. pp. 58-69
@inproceedings{38c7cfc112194c1499dd7e70b8ca0941,
title = "Persistence-based Interest Point Detection for 3D Deformable Surface",
abstract = "Several approaches for interest point detection on rigid shapes have been proposed, but few are available for non-rigid shapes. It is a very challenging task due to the presence of the large degrees of local deformations. This paper presents a novel method called persistence-based heat kernel signature (pHKS). It consists of two steps: scalar field construction and interest point detection. We propose to use the heat kernel signature function at a moderately small time scale to construct the scalar field. It has the advantage of being stable under various transformations. Based on the predefined scalar field, a 0-dimensional persistence diagram is computed, and the local geometric and global structural information of the shape are captured at the same time. Points with local maxima and high persistence are selected as interest points. We perform a comprehensive evaluation on two popular datasets (i.e., PHOTOMESH and Interest Points Dataset) to show the effectiveness of our method. Compared with existing techniques, our interest point detector achieves a superior performance in terms of repeatability and distinctiveness.",
author = "Xupeng Wang and Ferdous Sohel and Mohammed Bennamoun and Yulan Guo and Hang Lei",
year = "2017",
language = "English",
isbn = "9789897582240",
volume = "1",
pages = "58--69",
editor = "Claudio, {Ana Paula} and Bechmann, {Dominique } and Braz, {Jose }",
booktitle = "Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications",
publisher = "Scitepress",

}

Wang, X, Sohel, F, Bennamoun, M, Guo, Y & Lei, H 2017, Persistence-based Interest Point Detection for 3D Deformable Surface. in AP Claudio, D Bechmann & J Braz (eds), Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications . vol. 1, Scitepress, Portugal, pp. 58-69, 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, Porto, Portugal, 27/02/17.

Persistence-based Interest Point Detection for 3D Deformable Surface. / Wang, Xupeng; Sohel, Ferdous; Bennamoun, Mohammed; Guo, Yulan; Lei, Hang.

Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications . ed. / Ana Paula Claudio; Dominique Bechmann; Jose Braz. Vol. 1 Portugal : Scitepress, 2017. p. 58-69.

Research output: Chapter in Book/Conference paperConference paper

TY - GEN

T1 - Persistence-based Interest Point Detection for 3D Deformable Surface

AU - Wang, Xupeng

AU - Sohel, Ferdous

AU - Bennamoun, Mohammed

AU - Guo, Yulan

AU - Lei, Hang

PY - 2017

Y1 - 2017

N2 - Several approaches for interest point detection on rigid shapes have been proposed, but few are available for non-rigid shapes. It is a very challenging task due to the presence of the large degrees of local deformations. This paper presents a novel method called persistence-based heat kernel signature (pHKS). It consists of two steps: scalar field construction and interest point detection. We propose to use the heat kernel signature function at a moderately small time scale to construct the scalar field. It has the advantage of being stable under various transformations. Based on the predefined scalar field, a 0-dimensional persistence diagram is computed, and the local geometric and global structural information of the shape are captured at the same time. Points with local maxima and high persistence are selected as interest points. We perform a comprehensive evaluation on two popular datasets (i.e., PHOTOMESH and Interest Points Dataset) to show the effectiveness of our method. Compared with existing techniques, our interest point detector achieves a superior performance in terms of repeatability and distinctiveness.

AB - Several approaches for interest point detection on rigid shapes have been proposed, but few are available for non-rigid shapes. It is a very challenging task due to the presence of the large degrees of local deformations. This paper presents a novel method called persistence-based heat kernel signature (pHKS). It consists of two steps: scalar field construction and interest point detection. We propose to use the heat kernel signature function at a moderately small time scale to construct the scalar field. It has the advantage of being stable under various transformations. Based on the predefined scalar field, a 0-dimensional persistence diagram is computed, and the local geometric and global structural information of the shape are captured at the same time. Points with local maxima and high persistence are selected as interest points. We perform a comprehensive evaluation on two popular datasets (i.e., PHOTOMESH and Interest Points Dataset) to show the effectiveness of our method. Compared with existing techniques, our interest point detector achieves a superior performance in terms of repeatability and distinctiveness.

UR - http://www.scitepress.org/DigitalLibrary/ProceedingsDetails.aspx?ID=fRAMqyhb3Pw=&t=1

M3 - Conference paper

SN - 9789897582240

VL - 1

SP - 58

EP - 69

BT - Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications

A2 - Claudio, Ana Paula

A2 - Bechmann, Dominique

A2 - Braz, Jose

PB - Scitepress

CY - Portugal

ER -

Wang X, Sohel F, Bennamoun M, Guo Y, Lei H. Persistence-based Interest Point Detection for 3D Deformable Surface. In Claudio AP, Bechmann D, Braz J, editors, Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications . Vol. 1. Portugal: Scitepress. 2017. p. 58-69