SANet:空间注意力机制下的LiDAR点云实时语义分割方法

doi:10.13474/j.cnki.11-2246.2022.0321

测绘通报 ›› 2022, Vol. 0 ›› Issue (11): 32-38.doi: 10.13474/j.cnki.11-2246.2022.0321

SANet:空间注意力机制下的LiDAR点云实时语义分割方法

王玮琦¹, 游雄¹, 苏明占¹, 张蓝天², 周雪莹³, 赵耀¹

1. 信息工程大学, 河南郑州 450052;
2. 北京市遥感信息研究所, 北京 100192;
3. 北京理工大学, 北京 100081

收稿日期:2021-11-09 发布日期:2022-12-08
通讯作者: 游雄,E-mail:youarexiong@163.com
作者简介:王玮琦(1994-),男,博士生,主要研究方向为战场环境仿真与机器地图。E-mail:809741461@qq.com
基金资助:
国家自然科学基金(42130112;42171456);中原学者科学家工作室资助项目;国家重点研发计划(2017YFB0503500);国家自然科学基金青年基金(41801317)

SANet:real time semantic segmentation method of LiDAR point cloud based on spatial attention mechanism

WANG Weiqi¹, YOU Xiong¹, SU Mingzhan¹, ZHANG Lantian², ZHOU Xueying³, ZHAO Yao¹

1. Information Engineering University, Zhengzhou 450052, China;
2. Beijing Institute of Remote Sensing Information, Beijing 100192, China;
3. Beijing Institute of Technology, Beijing 100081, China

Received:2021-11-09 Published:2022-12-08

摘要/Abstract

摘要： 语义分割是智能机器人由感知智能迈向认知智能的重要基础,当前针对点云数据的语义分割方法存在实时性差、精度低等现象。本文系统分析了点云经球面投影所得的距离图像与自然图像的差异,为基于距离图像的实时语义分割网络设计提供了思路。通过分析发现,距离图像具有强空间相关性的特点,将强空间相关性与注意力机制相结合,提出基于空间注意力机制下的LiDAR点云实时语义分割方法SANet。该方法能够高效地聚合空间分布特征与上下文特征,且模型参数量较少,满足实时性的要求。在SemanticKITTI数据集上的试验表明,与其他优秀算法相比,SANet兼顾了实时性与准确性,显著提高了LiDAR点云语义分割的精度,可为自动驾驶及其他机器人应用领域提供辅助支撑。

关键词: 空间注意力, 点云语义分割, SemanticKITTI, 距离图像

Abstract: Semantic segmentation is an important basis for intelligent robots to move from perceptual intelligence to cognitive intelligence. The current semantic segmentation methods for point cloud data have poor real-time performance and low accuracy. In this article, we systematically analyze the difference between the range images generated by spherical projection of point cloud and common images, and provide ideas for the design of real-time semantic segmentation neural network. Through the analysis, we find that the range images have the characteristics of strong spatial correlation. This article combines the strong spatial correlation with attention mechanism, then proposes a real-time semantic segmentation method SANet based on spatial attention mechanism. SANet can efficiently aggregate spatial distribution features and context features. And the model parameters are less, which can meet the real-time requirements. Experiments on the SemanticKITTI dataset show that SANet has both good real-time performance and high accuracy compared with other excellent algorithms. The spatial attention mechanism proposed in this article significantly improves the accuracy of semantic segmentation of LiDAR point cloud by efficiently aggregating spatial distribution features and context features, which can provide auxiliary support for autonomous driving and other robot applications.

Key words: spatial attention, semantic segmentation of point cloud, SemanticKITTI, range image

中图分类号:

P237

王玮琦, 游雄, 苏明占, 张蓝天, 周雪莹, 赵耀. SANet:空间注意力机制下的LiDAR点云实时语义分割方法[J]. 测绘通报, 2022, 0(11): 32-38.

WANG Weiqi, YOU Xiong, SU Mingzhan, ZHANG Lantian, ZHOU Xueying, ZHAO Yao. SANet:real time semantic segmentation method of LiDAR point cloud based on spatial attention mechanism[J]. Bulletin of Surveying and Mapping, 2022, 0(11): 32-38.

参考文献

[1] 中国人工智能2.0发展战略研究项目组. 中国人工智能2.0发展战略研究[M]. 杭州:浙江大学出版社, 2018.
[2] 高俊. 图到用时方恨少, 重绘河山待后生:《测绘学报》60年纪念与前瞻[J]. 测绘学报, 2017, 46(10):1219-1225.
[3] MILIOTO A, VIZZO I, BEHLEY J, et al. RangeNet:fast and accurate LiDAR semantic segmentation[C]//Proceedings of 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). Macau:IEEE, 2019.
[4] XU Chenfeng, WU Bichen, WANG Zining, et al. SqueezeSegV3:spatially-adaptive convolution for efficient point-cloud segmentation[C]//Proceedings of the 16th European Conference on Computer Vision. New York,USA:ACM, 2020.
[5] GAO Biao, PAN Yancheng, LI Chengkun, et al. Are we hungry for 3D LiDAR data for semantic segmentation? A survey of datasets and methods[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(7):6063-6081.
[6] QI C R, YI L, SU H, et al. PointNet++:deep hierarchical feature learning on point sets in a metric space[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach, USA:[s.n.], 2017.
[7] ENGELMANN F, KONTOGIANNI T, HERMANS A, et al. Exploring spatial context for 3D semantic segmentation of point clouds[C]//Proceedings of 2017 IEEE International Conference on Computer Vision Workshops. Venice, Italy:IEEE, 2017.
[8] ENGELMANN F, KONTOGIANNI T, SCHULT J, et al. Know what your neighbors do:3D semantic segmentation of point clouds[M]//Lecture Notes in Computer Science. Cham:Springer International Publishing, 2019.
[9] LI Yangyan, BU Rui, SUN Mingchao, et al. PointCNN:convolution on x-transformed points[C]//Proceedings of the 32nd International Conference on Neural Information Processing Systems. New York,USA:ACM, 2018.
[10] HUANG Q G, WANG W Y, NEUMANN U. Recurrent slice networks for 3D segmentation of point clouds[C]//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA:[s.n.], 2018.
[11] SU Hang, JAMPANI V, SUN Deqing, et al. SPLATNet:sparse lattice networks for point cloud processing[C]//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT, USA:IEEE,2018.
[12] LANDRIEU L, SIMONOVSKY M. Large-scale point cloud semantic segmentation with superpoint graphs[C]//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT, USA:IEEE, 2018.
[13] LIU Fangyu, LI Shuaipeng, ZHANG Liqiang, et al. 3DCNN-DQN-RNN:a deep reinforcement learning framework for semantic parsing of large-scale 3D point clouds[C]//Proceedings of 2017 IEEE International Conference on Computer Vision. Venice, Italy:IEEE,2017.
[14] LAWIN F J, DANELLJAN M, TOSTEBERG P, et al. Deep projective 3D semantic segmentation[M]//Computer Analysis of Images and Patterns. Cham:Springer International Publishing, 2017.
[15] WU Bichen, WAN A, YUE Xiangyu, et al. SqueezeSeg:convolutional neural nets with recurrent CRF for real-time road-object segmentation from 3D LiDAR point cloud[C]//Proceedings of 2018 IEEE International Conference on Robotics and Automation. Brisbane, QLD, Australia:IEEE,2018.
[16] WU Bichen, ZHOU Xuanyu, ZHAO Sicheng, et al. SqueezeSegV2:improved model structure and unsupervised domain adaptation for road-object segmentation from a LiDAR point cloud[C]//Proceedings of 2019 International Conference on Robotics and Automation (ICRA). Montreal, QC, Canada:IEEE, 2019.
[17] HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA:IEEE,2016.
[18] BEHLEY J, GARBADE M, MILIOTO A, et al. SemanticKITTI:a dataset for semantic scene understanding of LiDAR sequences[C]//Proceedings of 2019 IEEE/CVF International Conference on Computer Vision (ICCV). Seoul, Korea (South):IEEE, 2019.
[19] AKSOY E E, BACI S, CAVDAR S. SalsaNet:fast road and vehicle segmentation in LiDAR point clouds for autonomous driving[C]//Proceedings of 2020 IEEE Intelligent Vehicles Symposium. Las Vegas, NV, USA:IEEE, 2020.
[20] CORTINHAL T, TZELEPIS G, AKSOY E E. SalsaNext:fast, uncertainty-aware semantic segmentation of LiDAR point clouds[C]//Proceedings of the 15th International Symposium on Advances in Visual Computing. New York,USA:ACM, 2020.

SANet:空间注意力机制下的LiDAR点云实时语义分割方法

SANet:real time semantic segmentation method of LiDAR point cloud based on spatial attention mechanism

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价