Research Article
A Novel Method for Semantic Segmentation on Lidar Point Clouds
@INPROCEEDINGS{10.1007/978-3-031-65126-7_32, author={Fei Wang and Liangtian Wan and Yan Zhu and Lu Sun and Xiaowei Zhao and Jianbo Zheng and Xianpeng Wang}, title={A Novel Method for Semantic Segmentation on Lidar Point Clouds}, proceedings={Quality, Reliability, Security and Robustness in Heterogeneous Systems. 19th EAI International Conference, QShine 2023, Shenzhen, China, October 8 -- 9, 2023, Proceedings, Part I}, proceedings_a={QSHINE}, year={2024}, month={8}, keywords={Semantic Segmentation Lidar Point Clouds Deep Learning}, doi={10.1007/978-3-031-65126-7_32} }- Fei Wang
Liangtian Wan
Yan Zhu
Lu Sun
Xiaowei Zhao
Jianbo Zheng
Xianpeng Wang
Year: 2024
A Novel Method for Semantic Segmentation on Lidar Point Clouds
QSHINE
Springer
DOI: 10.1007/978-3-031-65126-7_32
Abstract
Autonomous driving relies on multiple sensors, such as lidar and cameras, to perceive the surrounding environment and the vehicle’s own position. Among them, lidar point cloud segmentation is a crucial and challenging task for 3D scene understanding. In this paper, we propose a novel deep learning method RPNet for lidar point cloud segmentation that combines range image-based segmentation and point based segmentation. Our method extracts point cloud features from range images and predicts 3D point cloud labels from point clouds. The segmentation results of both branches are fused to improve accuracy. We evaluate our method on the Semantic KITTI dataset and show that it outperforms other fusion algorithms in terms of effectiveness and robustness.
