Skip to main content

Research Repository

Advanced Search

PointGS: Bridging and fusing geometric and semantic space for 3D point cloud analysis

Jiang, Chenru; Huang, Kaizhu; Wu, Junwei; Wang, Xinheng; Xiao, Jimin; Hussain, Amir

Authors

Chenru Jiang

Kaizhu Huang

Junwei Wu

Xinheng Wang

Jimin Xiao



Abstract

Directly processing 3D point cloud data becomes dominant in classification and segmentation tasks. Present mainstream point based methods usually focus on learning in either geometric space ( PointNet++) or semantic space ( DGCNN). Owing to the irregular and unordered data property of point cloud, these methods still suffer from drawbacks of either ambiguous local features aggregation in geometric space or poor global features extraction in semantic space. While few prior works address these two defects simultaneously by fusing information from the dual spaces, we make a first attempt to develop a synergistic framework, called PointGS. Leveraging both the strength of geometric structure and semantic representation, PointGS establishes a mutual supervision mechanism that can bridge the two spaces and fuse complementary information for better analyzing 3D point cloud data. Compared with existing popular networks, our work attains obvious performance improvement on all three mainstream tasks without any sophisticated operations.

Citation

Jiang, C., Huang, K., Wu, J., Wang, X., Xiao, J., & Hussain, A. (2023). PointGS: Bridging and fusing geometric and semantic space for 3D point cloud analysis. Information Fusion, 91, 316-326. https://doi.org/10.1016/j.inffus.2022.10.016

Journal Article Type Article
Acceptance Date Oct 15, 2022
Online Publication Date Oct 24, 2022
Publication Date 2023-03
Deposit Date Feb 13, 2023
Publicly Available Date Mar 28, 2024
Journal Information Fusion
Print ISSN 1566-2535
Publisher Elsevier
Peer Reviewed Peer Reviewed
Volume 91
Pages 316-326
DOI https://doi.org/10.1016/j.inffus.2022.10.016
Keywords Point cloud, Geometric space learning, Semantic space learning, Information fusion

Files

PointGS: Bridging And Fusing Geometric And Semantic Space For 3D Point Cloud Analysis (accepted version) (11.9 Mb)
PDF




You might also like



Downloadable Citations