A High-Speed and Computational Cost-Effective 3D Recognition Method With 2D-Edges-Based 4-Points Congruent Set Algorithm

被引：0

作者：

Shi, Xiaohang ^{[1
]}

Guo, Qitong ^{[1
]}

Murakami, Kenichi ^{[2
]}

Yamakawa, Yuji ^{[3
]}

机构：

[1] Univ Tokyo, Grad Sch Engn, Bunkyo City, 1138654, Japan

[2] Univ Tokyo, Inst Ind Sci, Bunkyo City 1138654, Japan

[3] Univ Tokyo, Interfac Initiat Informat Studies, Bunkyo City 1138654, Japan

来源：

IEEE ACCESS | 2024年 / 12卷

关键词：

Image edge detection; Three-dimensional displays; Computational modeling; Atmospheric modeling; Accuracy; Real-time systems; Pipelines; Parallel processing; Instance segmentation; Industrial robots; High-speed electronics; 3D object recognition; high-speed; robot;

D O I：

10.1109/ACCESS.2024.3475731

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

With the increasing application of robots in industrial manufacturing, the 3D vision perception is becoming more and more important. However, many researchers prefer to focus on the accuracy of 3D recognition rather than the speed. Faster speed can accelerate industrial production a lot. Meanwhile, the cost is also an important factor in reality. Since these, we propose the 2D-Edge-based 4-points congruent set (4PCS) algorithm. This algorithm utilizes the affine invariance and the projection invariance to find the correspondence between 3D model and 2D depth image, resulting in a significant increase in speed and keeping the accuracy. Its necessary hardware are just a consumer-level CPU and a consumer-level depth camera. On the Sil & eacute;ane Dataset, the average recognition time is 0.5ms per image.

引用

页码：151112 / 151121

页数：10

共 28 条

[1] 4-points congruent sets for robust pairwise surface registration [J].

Aiger, Dror ;

Mitra, Niloy J. ;

Cohen-Or, Daniel .

ACM TRANSACTIONS ON GRAPHICS, 2008, 27 (03)

[2] Defining the Pose of Any 3D Rigid Object and an Associated Distance [J].

Bregier, Romain ;

Devernay, Frederic ;

Leyrit, Laetitia ;

Crowley, James L. .

INTERNATIONAL JOURNAL OF COMPUTER VISION, 2018, 126 (06) :571-596

[3] Symmetry Aware Evaluation of 3D Object Detection and Pose Estimation in Scenes of Many Parts in Bulk [J].

Bregier, Romain ;

Devernay, Frederic ;

Leyrit, Laetitia ;

Crowley, James L. .

2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS (ICCVW 2017), 2017, :2209-2218

[4]

Chen K, 2022, P EUR C COMP VIS, P1

[5] G2L-Net: Global to Local Network for Real-time 6D Pose Estimation with Embedding Vector Features [J].

Chen, Wei ;

Jia, Xi ;

Chang, Hyung Jin ;

Duan, Jinming ;

Leonardis, Ales .

2020 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2020, :4232-4241

[6] A Comprehensive Study of 3-D Vision-Based Robot Manipulation [J].

Cong, Yang ;

Chen, Ronghan ;

Ma, Bingtao ;

Liu, Hongsen ;

Hou, Dongdong ;

Yang, Chenguang .

IEEE TRANSACTIONS ON CYBERNETICS, 2023, 53 (03) :1682-1698

[7] Model Globally, Match Locally: Efficient and Robust 3D Object Recognition [J].

Drost, Bertram ;

Ulrich, Markus ;

Navab, Nassir ;

Ilic, Slobodan .

2010 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2010, :998-1005

[8] Towards Robust Part-aware Instance Segmentation for Industrial Bin Picking [J].

Feng, Yidan ;

Yang, Biqi ;

Li, Xianzhi ;

Fu, Chi-Wing ;

Cao, Rui ;

Chen, Kai ;

Dou, Qi ;

Wei, Mingqiang ;

Liu, Yun-Hui ;

Heng, Pheng-Ann .

2022 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA 2022), 2022,

[9]

Gao G, 2020, IEEE INT CONF ROBOT, P3643, DOI [10.1109/ICRA40945.2020.9197461, 10.1109/icra40945.2020.9197461]

[10] Automatic markerless registration of point clouds with semantic-keypoint-based 4-points congruent sets [J].

Ge, Xuming .

ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING, 2017, 130 :344-357

← 1 2 3 →