Geometric Edge Description and Classification in Point Cloud Data with Application to 3D Object Recognition

Troels Bo Jørgensen, Anders Glent Buch, Dirk Kraft

Publikation: Bidrag til bog/antologi/rapport/konference-proceedingKonferencebidrag i proceedingsForskningpeer review

Resumé

This paper addresses the detection of geometric edges on 3D shapes. We investigate the use of local point cloud features and cast the edge detection problem as a learning problem. We show how supervised learning techniques can be applied to an existing shape description in terms of local feature descriptors. We apply our approach to several well-known shape descriptors. As an additional contribution, we develop a novel shape descriptor, termed
Equivalent Circumference Surface Angle Descriptor or ECSAD, which is particularly suitable for capturing local surface properties near edges. Our proposed descriptor allows for both fast computation and fast processing by having a low dimension, while still producing highly reliable edge detections. Lastly, we use our features in a 3D object recognition application using a well-established benchmark. We show that our edge features allow for significant speedups while achieving state of the art results.
OriginalsprogEngelsk
TitelProceedings of the 10th International Conference on Computer Vision Theory and Applications
RedaktørerJosé Braz, Sebastiano Battiato, Francisco Imai
Vol/bind1
ForlagInstitute for Systems and Technologies of Information, Control and Communication
Publikationsdato11. mar. 2015
Sider333-340
ISBN (Elektronisk)978-989-758-089-5
DOI
StatusUdgivet - 11. mar. 2015
Begivenhed10th International Conference on Computer Vision Theory and Applications - Berlin, Tyskland
Varighed: 11. mar. 201514. mar. 2015

Konference

Konference10th International Conference on Computer Vision Theory and Applications
LandTyskland
ByBerlin
Periode11/03/201514/03/2015

Fingeraftryk

Object recognition
Edge detection
Supervised learning
Surface properties
Processing

Citer dette

Jørgensen, T. B., Buch, A. G., & Kraft, D. (2015). Geometric Edge Description and Classification in Point Cloud Data with Application to 3D Object Recognition. I J. Braz, S. Battiato, & F. Imai (red.), Proceedings of the 10th International Conference on Computer Vision Theory and Applications (Bind 1, s. 333-340). Institute for Systems and Technologies of Information, Control and Communication. https://doi.org/10.5220/0005196703330340
Jørgensen, Troels Bo ; Buch, Anders Glent ; Kraft, Dirk. / Geometric Edge Description and Classification in Point Cloud Data with Application to 3D Object Recognition. Proceedings of the 10th International Conference on Computer Vision Theory and Applications. red. / José Braz ; Sebastiano Battiato ; Francisco Imai. Bind 1 Institute for Systems and Technologies of Information, Control and Communication, 2015. s. 333-340
@inproceedings{543c26e5297d411191e4f6e90f821f14,
title = "Geometric Edge Description and Classification in Point Cloud Data with Application to 3D Object Recognition",
abstract = "This paper addresses the detection of geometric edges on 3D shapes. We investigate the use of local point cloud features and cast the edge detection problem as a learning problem. We show how supervised learning techniques can be applied to an existing shape description in terms of local feature descriptors. We apply our approach to several well-known shape descriptors. As an additional contribution, we develop a novel shape descriptor, termedEquivalent Circumference Surface Angle Descriptor or ECSAD, which is particularly suitable for capturing local surface properties near edges. Our proposed descriptor allows for both fast computation and fast processing by having a low dimension, while still producing highly reliable edge detections. Lastly, we use our features in a 3D object recognition application using a well-established benchmark. We show that our edge features allow for significant speedups while achieving state of the art results.",
keywords = "Edge Detection, Object Recognition, Pose Estimation",
author = "J{\o}rgensen, {Troels Bo} and Buch, {Anders Glent} and Dirk Kraft",
year = "2015",
month = "3",
day = "11",
doi = "10.5220/0005196703330340",
language = "English",
volume = "1",
pages = "333--340",
editor = "Braz, {Jos{\'e} } and Battiato, {Sebastiano } and Imai, {Francisco }",
booktitle = "Proceedings of the 10th International Conference on Computer Vision Theory and Applications",
publisher = "Institute for Systems and Technologies of Information, Control and Communication",
address = "Portugal",

}

Jørgensen, TB, Buch, AG & Kraft, D 2015, Geometric Edge Description and Classification in Point Cloud Data with Application to 3D Object Recognition. i J Braz, S Battiato & F Imai (red), Proceedings of the 10th International Conference on Computer Vision Theory and Applications. bind 1, Institute for Systems and Technologies of Information, Control and Communication, s. 333-340, 10th International Conference on Computer Vision Theory and Applications, Berlin, Tyskland, 11/03/2015. https://doi.org/10.5220/0005196703330340

Geometric Edge Description and Classification in Point Cloud Data with Application to 3D Object Recognition. / Jørgensen, Troels Bo; Buch, Anders Glent; Kraft, Dirk.

Proceedings of the 10th International Conference on Computer Vision Theory and Applications. red. / José Braz; Sebastiano Battiato; Francisco Imai. Bind 1 Institute for Systems and Technologies of Information, Control and Communication, 2015. s. 333-340.

Publikation: Bidrag til bog/antologi/rapport/konference-proceedingKonferencebidrag i proceedingsForskningpeer review

TY - GEN

T1 - Geometric Edge Description and Classification in Point Cloud Data with Application to 3D Object Recognition

AU - Jørgensen, Troels Bo

AU - Buch, Anders Glent

AU - Kraft, Dirk

PY - 2015/3/11

Y1 - 2015/3/11

N2 - This paper addresses the detection of geometric edges on 3D shapes. We investigate the use of local point cloud features and cast the edge detection problem as a learning problem. We show how supervised learning techniques can be applied to an existing shape description in terms of local feature descriptors. We apply our approach to several well-known shape descriptors. As an additional contribution, we develop a novel shape descriptor, termedEquivalent Circumference Surface Angle Descriptor or ECSAD, which is particularly suitable for capturing local surface properties near edges. Our proposed descriptor allows for both fast computation and fast processing by having a low dimension, while still producing highly reliable edge detections. Lastly, we use our features in a 3D object recognition application using a well-established benchmark. We show that our edge features allow for significant speedups while achieving state of the art results.

AB - This paper addresses the detection of geometric edges on 3D shapes. We investigate the use of local point cloud features and cast the edge detection problem as a learning problem. We show how supervised learning techniques can be applied to an existing shape description in terms of local feature descriptors. We apply our approach to several well-known shape descriptors. As an additional contribution, we develop a novel shape descriptor, termedEquivalent Circumference Surface Angle Descriptor or ECSAD, which is particularly suitable for capturing local surface properties near edges. Our proposed descriptor allows for both fast computation and fast processing by having a low dimension, while still producing highly reliable edge detections. Lastly, we use our features in a 3D object recognition application using a well-established benchmark. We show that our edge features allow for significant speedups while achieving state of the art results.

KW - Edge Detection

KW - Object Recognition

KW - Pose Estimation

U2 - 10.5220/0005196703330340

DO - 10.5220/0005196703330340

M3 - Article in proceedings

VL - 1

SP - 333

EP - 340

BT - Proceedings of the 10th International Conference on Computer Vision Theory and Applications

A2 - Braz, José

A2 - Battiato, Sebastiano

A2 - Imai, Francisco

PB - Institute for Systems and Technologies of Information, Control and Communication

ER -

Jørgensen TB, Buch AG, Kraft D. Geometric Edge Description and Classification in Point Cloud Data with Application to 3D Object Recognition. I Braz J, Battiato S, Imai F, red., Proceedings of the 10th International Conference on Computer Vision Theory and Applications. Bind 1. Institute for Systems and Technologies of Information, Control and Communication. 2015. s. 333-340 https://doi.org/10.5220/0005196703330340