用于SLAM的点云动态物体识别

doi:10.3778/j.issn.1002-8331.2306-0225

摘要/Abstract

摘要： 检测和分割场景中动态物体对于建立一致性地图至关重要。针对当前点云动态物体检测算法依赖大量含有动态属性标注的数据、限制激光雷达扫描方式等问题，提出了一种基于连续点云的动态物体检测算法。将待预测点云、相邻帧点云以及通过SLAM（simultaneous localization and mapping）得到的位姿信息作为输入，利用点云场景流估计算法逐点估计移动情况，结合点云聚类、主成分分析（principal component analysis，PCA）等技术，整合场景流结果以获取实例级移动信息以判断物体的动态属性，并将点云语义分割作为判别点是否属于可移动类别的插件以提升动态物体识别精度。所提算法不需要具有动态属性的标注数据进行训练，并且对传感器的扫描方式、生成的点云数没有任何限制；与现有最先进的方法进行对比，具有易于训练、判断准确、结果鲁棒等特性。

关键词: 即时定位与地图构建（SLAM）, 深度学习, 点云

Abstract: Detecting and segmenting dynamic objects in a scene is crucial for building a consistent map. A dynamic object detection algorithm based on continuous point clouds is proposed to address the problems that current point cloud dynamic object detection algorithms rely on a large amount of data containing dynamic attribute annotations and limit the scanning method of LiDAR. The input for this algorithm consists of the point cloud to be predicted, adjacent frame point clouds, and pose information obtained through simultaneous localization and mapping (SLAM). A point cloud scene flow estimation algorithm is employed by this algorithm to estimate motion at a per-point level, and scene flow results are integrated using techniques such as point cloud clustering and principal component analysis (PCA) to infer instance-level motion information for the determination of dynamic object attributes. Furthermore, point cloud semantic segmentation is utilized as a plug-in to boost dynamic object recognition accuracy by categorizing points into movable categories. The need for annotated data with dynamic attributes for training is eliminated by the proposed algorithm, and it does not impose any restrictions on sensor scanning methods or the number of generated point clouds. Comparative analysis with state-of-the-art methods demonstrates the algorithm’s ease of training, high accuracy in object classification, and robustness.

Key words: simultaneous localization and mapping (SLAM), deep learning, point clouds

代领, 宋振波, 陆建峰. 用于SLAM的点云动态物体识别[J]. 计算机工程与应用, 2024, 60(20): 312-319.

DAI Ling, SONG Zhenbo, LU Jianfeng. Point Cloud Dynamic Object Recognition for SLAM[J]. Computer Engineering and Applications, 2024, 60(20): 312-319.

参考文献

[1] YOON D, TANG T, BARFOOT T. Mapless online detection of dynamic objects in 3D LiDAR[C]//Proceedings of the Conference on Computer and Robot Vision, 2019: 113-120.
[2] DEWAN A, CASELITZ T, TIPALDI G, et al. Rigid scene flow for 3D LiDAR scans[C]//Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, 2016: 1765-1770.
[3] KIM G, KIM A. Remove, then revert: static point cloud map construction using multiresolution range images[C]//Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, 2020: 10758-10765.
[4] 孙祥峻, 刘志刚. 矿井环境基于稀疏激光雷达数据的动态物体检测与追踪[J]. 工业控制计算机, 2020, 33(7): 91-93.
SUN X J, LIU Z G. Dynamic object detection and tracking based on sparse LiDAR data in mining environment[J]. Industrial Control Computer, 2020, 33(7): 91-93.
[5] DENG D, ZAKHOR A. RSF: optimizing rigid scene flow from 3D point clouds without labels[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2023: 1277-1286.
[6] LIN J, ZHANG F. Loam livox: a fast, robust, high-precision LiDAR odometry and mapping package for LiDARs of small FoV[C]//Proceedings of the 2020 IEEE International Conference on Robotics and Automation, 2020: 3126-3131.
[7] ZHOU Y, TUZEL O. VoxelNet: end-to-end learning for point cloud based 3D object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018: 4490-4499.
[8] LANG H, VORA S, CAESAR H, et al. PointPillars: fast encoders for object detection from point clouds[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 12689-12697.
[9] VAQUERO V, PINO I, MORENO F, et al. Dual-branch CNNs for vehicle detection and tracking on LiDAR data[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 22(11): 6942-6953.
[10] SHI S, GUO C, JIANG L, et al. PV-RCNN: point-voxel feature set abstraction for 3D object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 10526-10535.
[11] MENZE M, HEIPKE C, GEIGER A. Joint 3D estimation of vehicles and scene flow[J].ISPRS Annals of Photogrammetry, Remote Sensing & Spatial Information Sciences, 2015: 427-434.
[12] MENZE M, HEIPKE C, GEIGER A. Object scene flow[J].ISPRS Journal of Photogrammetry and Remote Sensing, 2018,140: 60-76.
[13] GEIGER A, LENZ P, URTASUN R. Are we ready for autonomous driving? the kitti vision benchmark suite[C]//Proceedings of the Conference on Computer Vision and Pattern Recognition, 2012: 3354-3361.
[14] BEHLEY J, GARBADE M, MILIOTO A, et al. SemanticKITTI: a dataset for semantic scene understanding of LiDAR sequences[C]//Proceedings of the IEEE International Conference on Computer Vision, 2019: 9296-9306.
[15] CHEN X, LI S, MERSCH B, et al. Moving object segmentation in 3D LiDAR data: a learning-based approach exploiting sequential data[J].IEEE Robotics and Automation Letters,2021, 6(4): 6529-6536.
[16] SUN J, DAI Y, ZHANG X, et al. Efficient spatial-temporal information fusion for LiDAR-based 3D moving object segmentation[C]//Proceedings of the 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2022: 11456-11463.
[17] YAN J, CHEN D, MYEONG H, et al. Automatic extraction of moving objects from image and LiDAR sequences[C]//Proceedings of the International Conference on 3D Vision, 2014: 673-680.
[18] POSTICA G, ROMANONI A, MATTEUCCI M. Robust moving objects detection in LiDAR data exploiting visual cues[C]//Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, 2016: 1093-1098.
[19] KITTENPLON Y, ELDAR Y, RAVIV D. FlowStep3D: model unrolling for self-supervised scene flow estimation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021: 4112-4121.
[20] 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, 2017: 5099-5108.
[21] XIN L, CHEN Y K, LU F, et al. Spherical transformer for LiDAR-based 3D recognition[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023: 17545-17555.
[22] 张冬冬, 郭杰, 陈阳. 基于原始点云的三维目标检测算法[J]. 计算机工程与应用, 2023, 59(3): 209-217.
ZHANG D D, GUO J, CHEN Y. 3D object detection algorithm based on raw point clouds[J]. Computer Engineering and Applications, 2023, 59(3): 209-217.
[23] LIU X, QI C R, GUIBAS J. FlowNet3D: learning scene flow in 3D point clouds[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 529-537.
[24] BAUR S, EMMERICHS J, MOOSMANN F, et al. SLIM: self-supervised LiDAR scene flow and motion segmentation[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021: 13106-13116.
[25] DENG Y, YANG J, XU D, et al. Multi-scale clustering by building a robust density map[C]//Proceedings of the European Conference on Computer Vision, 2017: 715-731.
[26] QI C R, KOLTUN V. PointNet: deep learning on point sets for 3D classification and segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 652-660.
[27] POWELL D, ABEL T. An exact general remeshing scheme applied to physically conservative voxelization[J].Journal of Computational Physics, 2015, 297: 340-356.
[28] CORTINHAL T, TZELEPIS G, AKSOY E. SalsaNext: fast, uncertainty-aware semantic segmentation of LiDAR point clouds[C]//Proceedings of the IEEE Vehicles Symposium, 2020: 207-222.
[29] XINGE Z, ZHOU H, WANG T, et al. Cylindrical and asymmetrical 3D convolution networks for LiDAR segmentation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021: 6807-6822.