MSIDA-Net: Point Cloud Semantic Segmentation via Multi-Spatial Information and Dual Adaptive Blocks

Large-scale 3D point clouds are rich in geometric shape and scale information but they are also scattered, disordered and unevenly distributed. These characteristics lead to difficulties in learning point cloud semantic segmentations. Although many works have performed well in this task, most of the...

Full description

Bibliographic Details
Main Authors: Feng Shuang, Pei Li, Yong Li, Zhenxin Zhang, Xu Li
Format: Article
Language:English
Published: MDPI AG 2022-05-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/14/9/2187
_version_ 1797502968424038400
author Feng Shuang
Pei Li
Yong Li
Zhenxin Zhang
Xu Li
author_facet Feng Shuang
Pei Li
Yong Li
Zhenxin Zhang
Xu Li
author_sort Feng Shuang
collection DOAJ
description Large-scale 3D point clouds are rich in geometric shape and scale information but they are also scattered, disordered and unevenly distributed. These characteristics lead to difficulties in learning point cloud semantic segmentations. Although many works have performed well in this task, most of them lack research on spatial information, which limits the ability to learn and understand the complex geometric structure of point cloud scenes. To this end, we propose the multispatial information and dual adaptive (MSIDA) module, which consists of a multispatial information encoding (MSI) block and dual adaptive (DA) blocks. The MSI block transforms the information of the relative position of each centre point and its neighbouring points into a cylindrical coordinate system and spherical coordinate system. Then the spatial information among the points can be re-represented and encoded. The DA blocks include a Coordinate System Attention Pooling Fusion (CSAPF) block and a Local Aggregated Feature Attention (LAFA) block. The CSAPF block weights and fuses the local features in the three coordinate systems to further learn local features, while the LAFA block weights the local aggregated features in the three coordinate systems to better understand the scene in the local region. To test the performance of the proposed method, we conducted experiments on the S3DIS, Semantic3D and SemanticKITTI datasets and compared the proposed method with other networks. The proposed method achieved 73%, 77.8% and 59.8% mean Intersection over Union (mIoU) on the S3DIS, Semantic3D and SemanticKITTI datasets, respectively.
first_indexed 2024-03-10T03:43:45Z
format Article
id doaj.art-3bf3e889d188474f9c092b3dfcc19f92
institution Directory Open Access Journal
issn 2072-4292
language English
last_indexed 2024-03-10T03:43:45Z
publishDate 2022-05-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj.art-3bf3e889d188474f9c092b3dfcc19f922023-11-23T09:11:56ZengMDPI AGRemote Sensing2072-42922022-05-01149218710.3390/rs14092187MSIDA-Net: Point Cloud Semantic Segmentation via Multi-Spatial Information and Dual Adaptive BlocksFeng Shuang0Pei Li1Yong Li2Zhenxin Zhang3Xu Li4Guangxi Key Laboratory of Intelligent Control and Maintenance of Power Equipment, School of Electrical Engineering, Guangxi University, Nanning 530004, ChinaGuangxi Key Laboratory of Intelligent Control and Maintenance of Power Equipment, School of Electrical Engineering, Guangxi University, Nanning 530004, ChinaGuangxi Key Laboratory of Intelligent Control and Maintenance of Power Equipment, School of Electrical Engineering, Guangxi University, Nanning 530004, ChinaBeijing Advanced Innovation Center for Imaging Theory and Technology, Key Lab of 3D Information Acquisition and Application, College of Resource Environment and Tourism, Capital Normal University, Beijing 100048, ChinaGuangxi Key Laboratory of Intelligent Control and Maintenance of Power Equipment, School of Electrical Engineering, Guangxi University, Nanning 530004, ChinaLarge-scale 3D point clouds are rich in geometric shape and scale information but they are also scattered, disordered and unevenly distributed. These characteristics lead to difficulties in learning point cloud semantic segmentations. Although many works have performed well in this task, most of them lack research on spatial information, which limits the ability to learn and understand the complex geometric structure of point cloud scenes. To this end, we propose the multispatial information and dual adaptive (MSIDA) module, which consists of a multispatial information encoding (MSI) block and dual adaptive (DA) blocks. The MSI block transforms the information of the relative position of each centre point and its neighbouring points into a cylindrical coordinate system and spherical coordinate system. Then the spatial information among the points can be re-represented and encoded. The DA blocks include a Coordinate System Attention Pooling Fusion (CSAPF) block and a Local Aggregated Feature Attention (LAFA) block. The CSAPF block weights and fuses the local features in the three coordinate systems to further learn local features, while the LAFA block weights the local aggregated features in the three coordinate systems to better understand the scene in the local region. To test the performance of the proposed method, we conducted experiments on the S3DIS, Semantic3D and SemanticKITTI datasets and compared the proposed method with other networks. The proposed method achieved 73%, 77.8% and 59.8% mean Intersection over Union (mIoU) on the S3DIS, Semantic3D and SemanticKITTI datasets, respectively.https://www.mdpi.com/2072-4292/14/9/2187LiDARpoint cloud dataspatial informationdeep learningsemantic segmentationconvolutional neural network
spellingShingle Feng Shuang
Pei Li
Yong Li
Zhenxin Zhang
Xu Li
MSIDA-Net: Point Cloud Semantic Segmentation via Multi-Spatial Information and Dual Adaptive Blocks
Remote Sensing
LiDAR
point cloud data
spatial information
deep learning
semantic segmentation
convolutional neural network
title MSIDA-Net: Point Cloud Semantic Segmentation via Multi-Spatial Information and Dual Adaptive Blocks
title_full MSIDA-Net: Point Cloud Semantic Segmentation via Multi-Spatial Information and Dual Adaptive Blocks
title_fullStr MSIDA-Net: Point Cloud Semantic Segmentation via Multi-Spatial Information and Dual Adaptive Blocks
title_full_unstemmed MSIDA-Net: Point Cloud Semantic Segmentation via Multi-Spatial Information and Dual Adaptive Blocks
title_short MSIDA-Net: Point Cloud Semantic Segmentation via Multi-Spatial Information and Dual Adaptive Blocks
title_sort msida net point cloud semantic segmentation via multi spatial information and dual adaptive blocks
topic LiDAR
point cloud data
spatial information
deep learning
semantic segmentation
convolutional neural network
url https://www.mdpi.com/2072-4292/14/9/2187
work_keys_str_mv AT fengshuang msidanetpointcloudsemanticsegmentationviamultispatialinformationanddualadaptiveblocks
AT peili msidanetpointcloudsemanticsegmentationviamultispatialinformationanddualadaptiveblocks
AT yongli msidanetpointcloudsemanticsegmentationviamultispatialinformationanddualadaptiveblocks
AT zhenxinzhang msidanetpointcloudsemanticsegmentationviamultispatialinformationanddualadaptiveblocks
AT xuli msidanetpointcloudsemanticsegmentationviamultispatialinformationanddualadaptiveblocks