A Proposal of 3D Feature Based on Occupancy of Point Cloud in Multiscale Shell Region

Shoichi Takei, Shuichi Akizuki, Manabu Hashimoto

Research output: Contribution to journalArticle

Abstract

In this paper, we propose a novel keypoint detection and feature description method called “SHORT” (Shell Histograms and Occupancy from Radial Transform) for fast 3D object recognition. Conventional keypoint detection and feature description methods such as the SHOT method have been necessary to calculate many normal vectors or other statistical values from the point cloud data in local regions, and therefore its computational costs are expensive. By contrast, the SHORT method consists of a fast keypoint detector that does not calculate statistics and a fast feature descriptor that uses a small number of points in the restricted local regions. The keypoint detector uses the occupancy measure which can be estimated by only counting the number of points in multiple spherical shell regions. Also, the feature descriptor uses a small number of points included in distinctive shell regions of multiple scales. Experimental results in 3D object recognition using real dataset show that the processing speed of the proposed method is approximately nine times faster than that of comparative methods.

Original languageEnglish
Pages (from-to)54-62
Number of pages9
JournalElectronics and Communications in Japan
Volume100
Issue number11
DOIs
Publication statusPublished - 2017 Nov 1
Externally publishedYes

Fingerprint

Point Cloud
Object recognition
histograms
proposals
Shell
Mathematical transformations
Detectors
spherical shells
detectors
3D Object Recognition
counting
Statistics
statistics
costs
Histogram
Descriptors
Processing
Detector
Transform
Costs

Keywords

  • 3D feature desription
  • 3D object recognition
  • multiple scale shell
  • occupancy
  • robot vision

ASJC Scopus subject areas

  • Signal Processing
  • Physics and Astronomy(all)
  • Computer Networks and Communications
  • Electrical and Electronic Engineering
  • Applied Mathematics

Cite this

A Proposal of 3D Feature Based on Occupancy of Point Cloud in Multiscale Shell Region. / Takei, Shoichi; Akizuki, Shuichi; Hashimoto, Manabu.

In: Electronics and Communications in Japan, Vol. 100, No. 11, 01.11.2017, p. 54-62.

Research output: Contribution to journalArticle

Takei, Shoichi ; Akizuki, Shuichi ; Hashimoto, Manabu. / A Proposal of 3D Feature Based on Occupancy of Point Cloud in Multiscale Shell Region. In: Electronics and Communications in Japan. 2017 ; Vol. 100, No. 11. pp. 54-62.
@article{7b3edb0acc1342a9babf2e8c4781796f,
title = "A Proposal of 3D Feature Based on Occupancy of Point Cloud in Multiscale Shell Region",
abstract = "In this paper, we propose a novel keypoint detection and feature description method called “SHORT” (Shell Histograms and Occupancy from Radial Transform) for fast 3D object recognition. Conventional keypoint detection and feature description methods such as the SHOT method have been necessary to calculate many normal vectors or other statistical values from the point cloud data in local regions, and therefore its computational costs are expensive. By contrast, the SHORT method consists of a fast keypoint detector that does not calculate statistics and a fast feature descriptor that uses a small number of points in the restricted local regions. The keypoint detector uses the occupancy measure which can be estimated by only counting the number of points in multiple spherical shell regions. Also, the feature descriptor uses a small number of points included in distinctive shell regions of multiple scales. Experimental results in 3D object recognition using real dataset show that the processing speed of the proposed method is approximately nine times faster than that of comparative methods.",
keywords = "3D feature desription, 3D object recognition, multiple scale shell, occupancy, robot vision",
author = "Shoichi Takei and Shuichi Akizuki and Manabu Hashimoto",
year = "2017",
month = "11",
day = "1",
doi = "10.1002/ecj.11994",
language = "English",
volume = "100",
pages = "54--62",
journal = "Electronics and Communications in Japan",
issn = "1942-9533",
publisher = "Scripta Technica",
number = "11",

}

TY - JOUR

T1 - A Proposal of 3D Feature Based on Occupancy of Point Cloud in Multiscale Shell Region

AU - Takei, Shoichi

AU - Akizuki, Shuichi

AU - Hashimoto, Manabu

PY - 2017/11/1

Y1 - 2017/11/1

N2 - In this paper, we propose a novel keypoint detection and feature description method called “SHORT” (Shell Histograms and Occupancy from Radial Transform) for fast 3D object recognition. Conventional keypoint detection and feature description methods such as the SHOT method have been necessary to calculate many normal vectors or other statistical values from the point cloud data in local regions, and therefore its computational costs are expensive. By contrast, the SHORT method consists of a fast keypoint detector that does not calculate statistics and a fast feature descriptor that uses a small number of points in the restricted local regions. The keypoint detector uses the occupancy measure which can be estimated by only counting the number of points in multiple spherical shell regions. Also, the feature descriptor uses a small number of points included in distinctive shell regions of multiple scales. Experimental results in 3D object recognition using real dataset show that the processing speed of the proposed method is approximately nine times faster than that of comparative methods.

AB - In this paper, we propose a novel keypoint detection and feature description method called “SHORT” (Shell Histograms and Occupancy from Radial Transform) for fast 3D object recognition. Conventional keypoint detection and feature description methods such as the SHOT method have been necessary to calculate many normal vectors or other statistical values from the point cloud data in local regions, and therefore its computational costs are expensive. By contrast, the SHORT method consists of a fast keypoint detector that does not calculate statistics and a fast feature descriptor that uses a small number of points in the restricted local regions. The keypoint detector uses the occupancy measure which can be estimated by only counting the number of points in multiple spherical shell regions. Also, the feature descriptor uses a small number of points included in distinctive shell regions of multiple scales. Experimental results in 3D object recognition using real dataset show that the processing speed of the proposed method is approximately nine times faster than that of comparative methods.

KW - 3D feature desription

KW - 3D object recognition

KW - multiple scale shell

KW - occupancy

KW - robot vision

UR - http://www.scopus.com/inward/record.url?scp=85030771669&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85030771669&partnerID=8YFLogxK

U2 - 10.1002/ecj.11994

DO - 10.1002/ecj.11994

M3 - Article

VL - 100

SP - 54

EP - 62

JO - Electronics and Communications in Japan

JF - Electronics and Communications in Japan

SN - 1942-9533

IS - 11

ER -