DGPolarNet: Dynamic Graph Convolution Network for LiDAR Point Cloud Semantic Segmentation on Polar BEV
Semantic segmentation in LiDAR point clouds has become an important research topic for autonomous driving systems. This paper proposes a dynamic graph convolution neural network for LiDAR point cloud semantic segmentation using a polar bird’s-eye view, referred to as DGPolarNet. LiDAR point clouds a...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-08-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/14/15/3825 |
_version_ | 1797412365308788736 |
---|---|
author | Wei Song Zhen Liu Ying Guo Su Sun Guidong Zu Maozhen Li |
author_facet | Wei Song Zhen Liu Ying Guo Su Sun Guidong Zu Maozhen Li |
author_sort | Wei Song |
collection | DOAJ |
description | Semantic segmentation in LiDAR point clouds has become an important research topic for autonomous driving systems. This paper proposes a dynamic graph convolution neural network for LiDAR point cloud semantic segmentation using a polar bird’s-eye view, referred to as DGPolarNet. LiDAR point clouds are converted to polar coordinates, which are rasterized into regular grids. The points mapped onto each grid distribute evenly to solve the problem of the sparse distribution and uneven density of LiDAR point clouds. In DGPolarNet, a dynamic feature extraction module is designed to generate edge features of perceptual points of interest sampled by the farthest point sampling and K-nearest neighbor methods. By embedding edge features with the original point cloud, local features are obtained and input into PointNet to quantize the points and predict semantic segmentation results. The system was tested on the Semantic KITTI dataset, and the segmentation accuracy reached 56.5% |
first_indexed | 2024-03-09T05:02:02Z |
format | Article |
id | doaj.art-68da0bfbc2c24b239d04445d603eaff3 |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-09T05:02:02Z |
publishDate | 2022-08-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-68da0bfbc2c24b239d04445d603eaff32023-12-03T12:59:17ZengMDPI AGRemote Sensing2072-42922022-08-011415382510.3390/rs14153825DGPolarNet: Dynamic Graph Convolution Network for LiDAR Point Cloud Semantic Segmentation on Polar BEVWei Song0Zhen Liu1Ying Guo2Su Sun3Guidong Zu4Maozhen Li5School of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaSchool of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaSchool of Information Science and Technology, North China University of Technology, Beijing 100144, ChinaDepartment of Computer and Information Technology, Purdue University, West Lafayette, IN 47907, USACOFCO Trading Agriculture & Big Data Solutions Co., Ltd., Dalian 116601, ChinaDepartment of Electronic and Electrical Engineering, Brunel University London, Uxbridge UB8 3PH, UKSemantic segmentation in LiDAR point clouds has become an important research topic for autonomous driving systems. This paper proposes a dynamic graph convolution neural network for LiDAR point cloud semantic segmentation using a polar bird’s-eye view, referred to as DGPolarNet. LiDAR point clouds are converted to polar coordinates, which are rasterized into regular grids. The points mapped onto each grid distribute evenly to solve the problem of the sparse distribution and uneven density of LiDAR point clouds. In DGPolarNet, a dynamic feature extraction module is designed to generate edge features of perceptual points of interest sampled by the farthest point sampling and K-nearest neighbor methods. By embedding edge features with the original point cloud, local features are obtained and input into PointNet to quantize the points and predict semantic segmentation results. The system was tested on the Semantic KITTI dataset, and the segmentation accuracy reached 56.5%https://www.mdpi.com/2072-4292/14/15/3825semantic segmentationpolar BEVLiDAR point clouddynamic graph convolution network |
spellingShingle | Wei Song Zhen Liu Ying Guo Su Sun Guidong Zu Maozhen Li DGPolarNet: Dynamic Graph Convolution Network for LiDAR Point Cloud Semantic Segmentation on Polar BEV Remote Sensing semantic segmentation polar BEV LiDAR point cloud dynamic graph convolution network |
title | DGPolarNet: Dynamic Graph Convolution Network for LiDAR Point Cloud Semantic Segmentation on Polar BEV |
title_full | DGPolarNet: Dynamic Graph Convolution Network for LiDAR Point Cloud Semantic Segmentation on Polar BEV |
title_fullStr | DGPolarNet: Dynamic Graph Convolution Network for LiDAR Point Cloud Semantic Segmentation on Polar BEV |
title_full_unstemmed | DGPolarNet: Dynamic Graph Convolution Network for LiDAR Point Cloud Semantic Segmentation on Polar BEV |
title_short | DGPolarNet: Dynamic Graph Convolution Network for LiDAR Point Cloud Semantic Segmentation on Polar BEV |
title_sort | dgpolarnet dynamic graph convolution network for lidar point cloud semantic segmentation on polar bev |
topic | semantic segmentation polar BEV LiDAR point cloud dynamic graph convolution network |
url | https://www.mdpi.com/2072-4292/14/15/3825 |
work_keys_str_mv | AT weisong dgpolarnetdynamicgraphconvolutionnetworkforlidarpointcloudsemanticsegmentationonpolarbev AT zhenliu dgpolarnetdynamicgraphconvolutionnetworkforlidarpointcloudsemanticsegmentationonpolarbev AT yingguo dgpolarnetdynamicgraphconvolutionnetworkforlidarpointcloudsemanticsegmentationonpolarbev AT susun dgpolarnetdynamicgraphconvolutionnetworkforlidarpointcloudsemanticsegmentationonpolarbev AT guidongzu dgpolarnetdynamicgraphconvolutionnetworkforlidarpointcloudsemanticsegmentationonpolarbev AT maozhenli dgpolarnetdynamicgraphconvolutionnetworkforlidarpointcloudsemanticsegmentationonpolarbev |