Integration method of local evidence for part-affordance estimation of everyday objects

Shuichi Akizuki, Masaki Iizuka, Kentaro Kozai, Manabu Hashimoto

研究成果: Article

抄録

Shapes of everyday objects are designed to achieve the predefined purpose of use. In this research, we call various kind of inherent functions as "part-affordance", and we have developed the method to perceive it from point cloud captured by a depth sensor. Difficulty of the estimation of it is that same local surfaces do not always have same affordance. In order to deal with this issue, we propose a method which integrates the evidence generated by local feature while considering the continuity of surface structure. Our experiments using a publicly available datasets confirmed that the proposed method have increased the recognition rate from 57% to 73% in comparison with the previous method. Moreover, we demonstrated that the robot arm can perform the task according to estimated part-affordances.

元の言語English
ページ(範囲)658-663
ページ数6
ジャーナルSeimitsu Kogaku Kaishi/Journal of the Japan Society for Precision Engineering
84
発行部数7
DOI
出版物ステータスPublished - 2018 1 1

Fingerprint

Surface structure
Robots
Sensors
Experiments

ASJC Scopus subject areas

  • Mechanical Engineering

これを引用

Integration method of local evidence for part-affordance estimation of everyday objects. / Akizuki, Shuichi; Iizuka, Masaki; Kozai, Kentaro; Hashimoto, Manabu.

:: Seimitsu Kogaku Kaishi/Journal of the Japan Society for Precision Engineering, 巻 84, 番号 7, 01.01.2018, p. 658-663.

研究成果: Article

Akizuki, Shuichi ; Iizuka, Masaki ; Kozai, Kentaro ; Hashimoto, Manabu. / Integration method of local evidence for part-affordance estimation of everyday objects. :: Seimitsu Kogaku Kaishi/Journal of the Japan Society for Precision Engineering. 2018 ; 巻 84, 番号 7. pp. 658-663.
@article{2a556652fe954b4fa7c406f22166c250,
title = "Integration method of local evidence for part-affordance estimation of everyday objects",
abstract = "Shapes of everyday objects are designed to achieve the predefined purpose of use. In this research, we call various kind of inherent functions as {"}part-affordance{"}, and we have developed the method to perceive it from point cloud captured by a depth sensor. Difficulty of the estimation of it is that same local surfaces do not always have same affordance. In order to deal with this issue, we propose a method which integrates the evidence generated by local feature while considering the continuity of surface structure. Our experiments using a publicly available datasets confirmed that the proposed method have increased the recognition rate from 57{\%} to 73{\%} in comparison with the previous method. Moreover, we demonstrated that the robot arm can perform the task according to estimated part-affordances.",
keywords = "3D object recognition, Affordance, Local evidence, Point cloud processing",
author = "Shuichi Akizuki and Masaki Iizuka and Kentaro Kozai and Manabu Hashimoto",
year = "2018",
month = "1",
day = "1",
doi = "10.2493/jjspe.84.658",
language = "English",
volume = "84",
pages = "658--663",
journal = "Seimitsu Kogaku Kaishi/Journal of the Japan Society for Precision Engineering",
issn = "0912-0289",
publisher = "Japan Society for Precision Engineering",
number = "7",

}

TY - JOUR

T1 - Integration method of local evidence for part-affordance estimation of everyday objects

AU - Akizuki, Shuichi

AU - Iizuka, Masaki

AU - Kozai, Kentaro

AU - Hashimoto, Manabu

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Shapes of everyday objects are designed to achieve the predefined purpose of use. In this research, we call various kind of inherent functions as "part-affordance", and we have developed the method to perceive it from point cloud captured by a depth sensor. Difficulty of the estimation of it is that same local surfaces do not always have same affordance. In order to deal with this issue, we propose a method which integrates the evidence generated by local feature while considering the continuity of surface structure. Our experiments using a publicly available datasets confirmed that the proposed method have increased the recognition rate from 57% to 73% in comparison with the previous method. Moreover, we demonstrated that the robot arm can perform the task according to estimated part-affordances.

AB - Shapes of everyday objects are designed to achieve the predefined purpose of use. In this research, we call various kind of inherent functions as "part-affordance", and we have developed the method to perceive it from point cloud captured by a depth sensor. Difficulty of the estimation of it is that same local surfaces do not always have same affordance. In order to deal with this issue, we propose a method which integrates the evidence generated by local feature while considering the continuity of surface structure. Our experiments using a publicly available datasets confirmed that the proposed method have increased the recognition rate from 57% to 73% in comparison with the previous method. Moreover, we demonstrated that the robot arm can perform the task according to estimated part-affordances.

KW - 3D object recognition

KW - Affordance

KW - Local evidence

KW - Point cloud processing

UR - http://www.scopus.com/inward/record.url?scp=85049781464&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85049781464&partnerID=8YFLogxK

U2 - 10.2493/jjspe.84.658

DO - 10.2493/jjspe.84.658

M3 - Article

AN - SCOPUS:85049781464

VL - 84

SP - 658

EP - 663

JO - Seimitsu Kogaku Kaishi/Journal of the Japan Society for Precision Engineering

JF - Seimitsu Kogaku Kaishi/Journal of the Japan Society for Precision Engineering

SN - 0912-0289

IS - 7

ER -