LSPConv: local spatial projection convolution for point cloud analysis
Authors:
Open Access
- 5 January 2024
- journal article
- research article
- Published by PeerJ in PeerJ Computer Science
- Vol. 10, e1738
Abstract
This study introduces a novel approach, Local Spatial Projection Convolution (LSPConv), for point cloud classification and semantic segmentation. Unlike conventional methods utilizing relative coordinates for local geometric information, our motivation stems from the inadequacy of existing techniques for representing the intricate spatial organization of unconsolidated and irregular 3D point clouds. To address this limitation, we propose a Local Spatial Projection Module utilizing a vector projection strategy, designed to capture comprehensive local spatial information more effectively. Moreover, recent studies emphasize the importance of anisotropic kernels for point cloud feature extraction, considering the distinct contributions of individual neighboring points. To cater to this requirement, we introduce the Feature Weight Assignment (FWA) Module to assign weights to neighboring points, enhancing the anisotropy crucial for accurate feature extraction. Additionally, we introduce an Anisotropic Relative Feature Encoding Module that adaptively encodes points based on their relative features, further amplifying the anisotropic characteristics. Our approaches achieve remarkable results for point cloud classification and segmentation in several benchmark datasets based on extensive qualitative and quantitative evaluation.Keywords
Funding Information
- National key research and development program (2022YFB4702402)
- National Natural Science Foundation of China (62176072)
- Anhui Province Key Laboratory of Machine Vision Inspection (KLMVI-2023-HIT-12)
This publication has 20 references indexed in Scilit:
- A Rotation-Invariant Framework for Deep Point Cloud AnalysisIEEE Transactions on Visualization and Computer Graphics, 2021
- PCT: Point cloud transformerComputational Visual Media, 2021
- Geometric Back-Projection Network for Point Cloud ClassificationIEEE Transactions on Multimedia, 2021
- PRA-Net: Point Relation-Aware Network for 3D Point Cloud AnalysisIEEE Transactions on Image Processing, 2021
- Learning of 3D Graph Convolution Networks for Point Cloud AnalysisIEEE Transactions on Pattern Analysis and Machine Intelligence, 2021
- Navigation of a mobile robot in a dynamic environment using a point cloud mapArtificial Life and Robotics, 2020
- Spherical Kernel for Efficient Graph Convolution on 3D Point CloudsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2020
- Dynamic Graph CNN for Learning on Point CloudsACM Transactions on Graphics, 2019
- O-CNNACM Transactions on Graphics, 2017
- A scalable active framework for region annotation in 3D shape collectionsACM Transactions on Graphics, 2016