Curvature-Variation-Inspired Sampling for Point Cloud Classification and Segmentation

Zhu，Lei1; Chen，Weinan1; Lin，Xubin1; He，Li2; Guan，Yisheng1

2022

10.1109/LSP.2022.3200585

IEEE SIGNAL PROCESSING LETTERS   Impact Factor & Quartile

1070-9908

1558-2361

Point cloud is a discrete and unordered expression of 3D data. A lot of methods have been proposed to solve the problem in 3D object classification and scene recognition. To handle the huge amount of unordered point cloud, down-sampling before processing is needed. The shortage of existing sampling methods is the lack of geometry information consideration, which is essential for point cloud classification and segmentation tasks. Our method is mainly motivated by the observation that points with a high curvature variation can depict the outlines of objects. Thus, we propose a curvature variation based sampling method for point cloud classification and segmentation tasks. We aim to sample points with high curvature variations, which are considered to be more suitable for classification and segmentation tasks than the traditional sampling method. We combine the proposed sampling algorithm with the existing sampling method for multiple information fusion, and a higher accuracy and mean IoU can be achieved. The experimental results verify the advantage of considering curvature variation in classification and segmentation tasks.

classification and segmentation Convolution curvature variation Geometry Point cloud Point cloud compression Sampling methods Shape Task analysis Three-dimensional displays

SCI

English

Others

Leading Talents of Guangdong Province Program["2016LJ06G498","2019QN01X761"] ; Program for Guangdong Yangfan Innovative and Entrepreneurial Teams[2017YT05G026] ; Guangdong Provincial Special Fund for Modern Agriculture Common Key Technology R&D Innovation Team[2019KJ129] ; China Postdoctoral Science Foundation[2021M701576] ; National Natural Science Foundation of China["62103179","62173096"]

Engineering

Engineering, Electrical & Electronic

WOS:000852825500004

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC

ENGINEERING

2-s2.0-85137551614

Scopus

1.Biomimetic and Intelligent Robotics Lab (BIRL), Guangdong University of Technology, Guangzhou, China
2.Department of Electronic and Electrical Engineering, Southern University of Science and Technology, Shenzhen, China

Zhu，Lei,Chen，Weinan,Lin，Xubin,et al. Curvature-Variation-Inspired Sampling for Point Cloud Classification and Segmentation[J]. IEEE SIGNAL PROCESSING LETTERS,2022,29:1868-1872.

Zhu，Lei,Chen，Weinan,Lin，Xubin,He，Li,&Guan，Yisheng.(2022).Curvature-Variation-Inspired Sampling for Point Cloud Classification and Segmentation.IEEE SIGNAL PROCESSING LETTERS,29,1868-1872.

Zhu，Lei,et al."Curvature-Variation-Inspired Sampling for Point Cloud Classification and Segmentation".IEEE SIGNAL PROCESSING LETTERS 29(2022):1868-1872.