Position and pose recognition of randomly stacked objects using highly observable 3D vector pairs

Shuichi Akizuki, Manabu Hashimoto

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

1 Citation (Scopus)

Abstract

We propose a fast and reliable 3D object detection method that can be applied for complicated scenes consisting of randomly stacked objects. The proposed method uses '3D vector pair' that has a common start point and different end points and it has surface normal distribution as the feature descriptor. By considering the observability of vector pairs, the proposed method has been achieved high recognition performance. Observability factor of the vector pair is calculated by simulating the visible state of the vector pair from various viewpoints. By integrating the observability factor and the distinctiveness factor proposed in our previous work, vector pairs that have effectiveness for matching are extracted and these are used for object pose estimation. Experiments have confirmed that the proposed method increases the recognition success rate from 45.8% to 93.1%, in comparison with the state-of-the-arts method. The processing time of the proposed method is fast enough for the robotic bin-picking.

Original languageEnglish
Title of host publicationIECON Proceedings (Industrial Electronics Conference)
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages5266-5271
Number of pages6
ISBN (Electronic)9781479940325
DOIs
Publication statusPublished - 2014 Feb 24
Externally publishedYes

Fingerprint

Observability
Bins
Normal distribution
Robotics
Processing
Experiments

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Akizuki, S., & Hashimoto, M. (2014). Position and pose recognition of randomly stacked objects using highly observable 3D vector pairs. In IECON Proceedings (Industrial Electronics Conference) (pp. 5266-5271). [7049303] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IECON.2014.7049303

Position and pose recognition of randomly stacked objects using highly observable 3D vector pairs. / Akizuki, Shuichi; Hashimoto, Manabu.

IECON Proceedings (Industrial Electronics Conference). Institute of Electrical and Electronics Engineers Inc., 2014. p. 5266-5271 7049303.

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

Akizuki, S & Hashimoto, M 2014, Position and pose recognition of randomly stacked objects using highly observable 3D vector pairs. in IECON Proceedings (Industrial Electronics Conference)., 7049303, Institute of Electrical and Electronics Engineers Inc., pp. 5266-5271. https://doi.org/10.1109/IECON.2014.7049303
Akizuki S, Hashimoto M. Position and pose recognition of randomly stacked objects using highly observable 3D vector pairs. In IECON Proceedings (Industrial Electronics Conference). Institute of Electrical and Electronics Engineers Inc. 2014. p. 5266-5271. 7049303 https://doi.org/10.1109/IECON.2014.7049303
Akizuki, Shuichi ; Hashimoto, Manabu. / Position and pose recognition of randomly stacked objects using highly observable 3D vector pairs. IECON Proceedings (Industrial Electronics Conference). Institute of Electrical and Electronics Engineers Inc., 2014. pp. 5266-5271
@inproceedings{b2a86ebb9d914fb2ab788b3f0b97504c,
title = "Position and pose recognition of randomly stacked objects using highly observable 3D vector pairs",
abstract = "We propose a fast and reliable 3D object detection method that can be applied for complicated scenes consisting of randomly stacked objects. The proposed method uses '3D vector pair' that has a common start point and different end points and it has surface normal distribution as the feature descriptor. By considering the observability of vector pairs, the proposed method has been achieved high recognition performance. Observability factor of the vector pair is calculated by simulating the visible state of the vector pair from various viewpoints. By integrating the observability factor and the distinctiveness factor proposed in our previous work, vector pairs that have effectiveness for matching are extracted and these are used for object pose estimation. Experiments have confirmed that the proposed method increases the recognition success rate from 45.8{\%} to 93.1{\%}, in comparison with the state-of-the-arts method. The processing time of the proposed method is fast enough for the robotic bin-picking.",
author = "Shuichi Akizuki and Manabu Hashimoto",
year = "2014",
month = "2",
day = "24",
doi = "10.1109/IECON.2014.7049303",
language = "English",
pages = "5266--5271",
booktitle = "IECON Proceedings (Industrial Electronics Conference)",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Position and pose recognition of randomly stacked objects using highly observable 3D vector pairs

AU - Akizuki, Shuichi

AU - Hashimoto, Manabu

PY - 2014/2/24

Y1 - 2014/2/24

N2 - We propose a fast and reliable 3D object detection method that can be applied for complicated scenes consisting of randomly stacked objects. The proposed method uses '3D vector pair' that has a common start point and different end points and it has surface normal distribution as the feature descriptor. By considering the observability of vector pairs, the proposed method has been achieved high recognition performance. Observability factor of the vector pair is calculated by simulating the visible state of the vector pair from various viewpoints. By integrating the observability factor and the distinctiveness factor proposed in our previous work, vector pairs that have effectiveness for matching are extracted and these are used for object pose estimation. Experiments have confirmed that the proposed method increases the recognition success rate from 45.8% to 93.1%, in comparison with the state-of-the-arts method. The processing time of the proposed method is fast enough for the robotic bin-picking.

AB - We propose a fast and reliable 3D object detection method that can be applied for complicated scenes consisting of randomly stacked objects. The proposed method uses '3D vector pair' that has a common start point and different end points and it has surface normal distribution as the feature descriptor. By considering the observability of vector pairs, the proposed method has been achieved high recognition performance. Observability factor of the vector pair is calculated by simulating the visible state of the vector pair from various viewpoints. By integrating the observability factor and the distinctiveness factor proposed in our previous work, vector pairs that have effectiveness for matching are extracted and these are used for object pose estimation. Experiments have confirmed that the proposed method increases the recognition success rate from 45.8% to 93.1%, in comparison with the state-of-the-arts method. The processing time of the proposed method is fast enough for the robotic bin-picking.

UR - http://www.scopus.com/inward/record.url?scp=84949929098&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949929098&partnerID=8YFLogxK

U2 - 10.1109/IECON.2014.7049303

DO - 10.1109/IECON.2014.7049303

M3 - Conference contribution

AN - SCOPUS:84949929098

SP - 5266

EP - 5271

BT - IECON Proceedings (Industrial Electronics Conference)

PB - Institute of Electrical and Electronics Engineers Inc.

ER -