Point Cloud Data Classification Method of PointNet++ with Position Adaptive Convolution

doi:10.3969/j.issn.1006-2475.2025.01.008

Abstract

Abstract: Aiming at the problem of low classification accuracy of point cloud data in complex scenes， a PointNet++ deep neural network model based on position adaptive convolution is proposed. Since adaptive position convolution has strong ability to capture fine-grained local features and can fully obtain the spatial variation and geometric structure feature information of three-dimensional point clouds， on the basis of PointNet++ network， the proposed method in this paper first obtains the key points through the farthest point sampling， and then uses the K nearest neighbor method to realize grouping according to the key points， and then using the adaptive position convolution replaces the MLP in the original method to extract the local features of each group， and finally completes the point cloud classification. The proposed method was compared on two public point cloud datasets S3DIS and Semantic3D. Experimental results show that the overall accuracy and mIoU of the proposed method on the indoor dataset S3DIS are about 2.7 percentage points and 3.2 percentage points higher than PointNet++ network， respectively， and the overall accuracy and mIoU of the outdoor dataset Semantic3D are about 2.5 percentage points and 2.1 percentage points higher than PointNet++.

Key words: , point cloud classification, position adaptive convolution, PointNet++, deep learning, local feature

CLC Number:

TP391

YAN Xiaoqi, PENG Yiqing, REN Xiaoling. Point Cloud Data Classification Method of PointNet++ with Position Adaptive Convolution[J]. Computer and Modernization, 2025, 0(01): 44-49.

References

［1］ DOU M， TAYLOR J， FUCHS H， et al. 3D scanning deformable objects with a single RGBD sensor［C］// 2015 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）. IEEE， 2015：493-501.
［2］ ZERMAS D， IZZAT I， PAPANILOLOPOULOS N. Fast segmentation of 3D point clouds：A paradigm on LiDAR data for auto nomous vehicle applications［C］// 2017 IEEE International Conference on Robotics and Automation （ICRA）. IEEE， 2017：5067-5073.
［3］ POLEWSKI P， YAO W， HEURICH M， et al. Detection of fallen trees in ALS point clouds using a normalized cut approach trained by simulation［J］. ISPRS Journal of Photogrammetry and Remote Sensing， 2015，105（7）：252-271.
［4］ YANG B S， HANG R G， LI J P， et al. Automated reconstruction of building LoDs from airborne LiDAR point clouds using an improved morphological scale space［J］. Remote Sensing， 2016，9（1）. DOI： 10.3390/rs9010014.
［5］ PAN Y， DONG Y Q， WANG D L， et al. Threedimensional reconstruction of structural surface model of heritage bridges using UAV-based photogrammetric point clouds［J］. Remote Sensing， 2019，11（10）. DOI： 10.3390/rs1110
1204.
［6］ KIM K， KIM C， JANG C， et al. Deep learning-based dynamic object classification using LiDAR point cloud au gmented by layer-based accumulation for intelligent vehicles［J］.Expert Systems with Applications， 2020，167. DOI： 10.1016/j.eswa.2020.113861.
［7］胡海瑛，惠振阳，李娜. 基于多基元特征向量融合的机载LiDAR点云分类［J］. 中国激光， 2020，47（8）：229-239.
［8］ COLGAN M S， BALDECK C A， FERET J B， et al. Mapping Savanna tree species at ecosystem scales using support vector machine classification and BRDF correctionon airborne hyperspectral and LiDAR data［J］. Remote Sensing， 2012，4（11）：3462-3480.
［9］段广明. 基于随机森林的机载激光雷达点云数据分类研究［D］. 西安：西安电子科技大学， 2018.
［10］周梦蝶. 融合影像匹配点云的机载 LiDAR 点云分类［D］. 北京：中国地质大学， 2020.
［11］ YU T， MENGJ J， YUAN J S. Multi-view harmonized bilinear network for 3D object recognition［C］// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. IEEE， 2018：186-194.
［12］ QI C R， SU H， MO K C， et al. PointNet： Deep learning on point sets for 3D classification and segmentation［J］// 2017 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）. IEEE， 2017：77-85.
［13］ QI C R， YI L， SU H， et al. PointNet++： Deep Hierarchical feature learning on point sets in a metric space［J］// Proceedings of the 31st International Conference on Neural Information Processing Systems. ACM， 2017：5105-5114.
［14］ JIANG M Y， WU Y R， ZHAO T Q， et al. PointSIFT： A SIFT-like network module for 3D point cloud semantic segmentation ［J］. arXiv preprint arXiv： 1807.00652， 2018.
［15］ LI J X， CHEN B M， LEE G H.SO-Net： Self-organizing network for point cloud analysis［J］// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）. IEEE， 2018：9397-9406.
［16］ LI Y Y， BU R， SUN M C， et al. PointCNN： Convolution on Χ-transformed points［C］// Proceedings of the 32nd International Conference on Neural Information Processing Systems. ACM， 2018：828-838.
［17］ GROH F， WIESCHOLLEK P， LENSCH H P A. Flex-convolution （deep learning beyond grid-worlds）［J］. arXiv preprint arXiv：1803.07289， 2018.
［18］ LIU Y C， FAN B， XIANG S M， et al. Relation-shape convolutional neural network for point cloud analysis［C］// Proceedings of the IEEE/CVF Conference Computer Vision and Pattern Recognition. IEEE， 2019：8895-8904.
［19］ SHEN Y R， FENG C， YANG Y Q， et al. Mining point cloud local structures by kernel correlation and graph pooling// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition （CVPR）. IEEE， 2018：4548-4557.
［20］ XU M T， DING R Y， ZHAO H S， et al. PAConv： Position adaptive convolution with dynamic kernel assembling on point clouds［J］// 2021 IEEE Conference on Computer Vision and Pattern Recognition （CVPR）. IEEE， 2021：3172-3181.
［21］田茂义，李文君. 一种基于改进最远点采样的点云数据处理方法：202310530587［P］. 2023-08-11.
［22］ LIN W S， FAN W W， LIU H R， et al. Classification of handheld laser scanning tree point cloud based on different KNN algorithms and random forest algorithm［J］. Forests， 2021，12（3）. DOI： 10.3390/f12030292.
［23］ ARMENI I， SAX S， ZAMIR A R， et al. Joint 2D-3D-Semantic data for indoor scene understanding［J］. arXiv preprint arXiv：1702.01105， 2017.
［24］ HACKEL T， SAVINOV N， LADICKY L， et al. Semantic3D.net： A new large-scale point cloud classification benchmark ［J］. arXiv preprint arXiv：1704.03847， 2017.
［25］ GARCIA-GARCIA A， ORTS-ESCOLANO S， OPREA S， et al. A survey on deep learning techniques for image video semantic segmentation［J］. Applied Soft Computing， 2018，70：41-65.
［26］ BOULCH A， GUERRY J， LE SAUX B， et al. SnapNet：3D point cloud semantic labeling with 2D deep segmantation networks［J］. Computers & Graphics， 2018，71：189-198.
［27］ TCHAPMI L， CHOY C， ARMENI I， et al. SEGCloud： Semantic segmentation of 3D point clouds［C］// Proceedings of the International Conference on 3D Vision. IEEE， 2017：537-547.

[1]	HAN Xiaolong, ZENG Xi, LIU Kun, SHANG Yu. Stance Detection with LoRA-based Fine-tuning General Language Model [J]. Computer and Modernization, 2025, 0(01): 1-6.
[2]	XU Shengchao, CHEN Fuqiang. A Food Safety Risk Warning Method Based on BP Neural Network [J]. Computer and Modernization, 2025, 0(01): 20-24.
[3]	ZHANG Yue, LI Huayu, ZHANG Zhikang, SHEN Xinyi. Academic Recommendation System Based on Knowledge Graph and Semantic Information [J]. Computer and Modernization, 2025, 0(01): 50-58.
[4]	XIE Zetao, ZHUANG Yi . Transient Fault Detection for Low-orbit Internet Communication System Based on CatBoost [J]. Computer and Modernization, 2025, 0(01): 59-66.
[5]	LIU Haitao, FENG Fan. BD Based Lattice Reduction Assisted Continuous Interference Cancellation Detection Algorithm [J]. Computer and Modernization, 2025, 0(01): 67-73.
[6]	LI Qiusheng1, LI Wei1, ZHOU Dongxu1, GUO Chuang1, JIN Wenxin 2. Optical Fiber Timing Synchronization in Power Systems Based on ARIMA-ELM Algorithm [J]. Computer and Modernization, 2025, 0(01): 74-79.
[7]	XIAO Junbi, FU Tianqi. Real-Time Traffic Classification Method Based on High-dimensional Feature#br# Dimensionality Reduction and Clustering [J]. Computer and Modernization, 2025, 0(01): 80-85.
[8]	LI Xi, PAN Yu. Ground Penetrating Radar Pipeline Object Detection Method Based on Improved YOLOv8 [J]. Computer and Modernization, 2025, 0(01): 94-99.
[9]	TU Fuquan, QI Yanqi, LIU Jian, WANG Shufeng. Metal Gear Surface Defect Detection Algorithm Based on Improved YOLOv8s [J]. Computer and Modernization, 2025, 0(01): 100-106.
[10]	GUO Xiaofeng, ZHUANG Yi. Randomized Spanning Tree Based Redundant Fault Tolerance Algorithm for SDN Control Plane [J]. Computer and Modernization, 2025, 0(01): 107-112.
[11]	ZHENG Jiuchao, ZHAO Xinyuan. Entity Linking Method Based on Topics and Description Information [J]. Computer and Modernization, 2024, 0(12): 10-14.
[12]	ZHAO Chenyang, XUE Tao, LIU Junhua. Fashion Clothing Pattern Generation Based on Improved Stable Diffusion [J]. Computer and Modernization, 2024, 0(12): 15-23.
[13]	WANG Xiaohang1, LI Yongjie1, YU Lei1, FAN Xiao2. A Method of Using Compound Event Probability Operation to Solve Problem of Negative Information Blocking Maximization [J]. Computer and Modernization, 2024, 0(12): 24-33.
[14]	ZHANG Xiaodong1, BAI Guangzhi1, LI Min1, LI Haoyang2. Oil and Gas Well Production Prediction Model Based on Empirical Wavelet Transform [J]. Computer and Modernization, 2024, 0(12): 53-58.
[15]	LIU Yunhai1, Feng Guang1, WU Xiaoting2, YANG Qun2 . Safety Helmet Wearing Detection Algorithm for Complex Construction Scenes [J]. Computer and Modernization, 2024, 0(12): 66-71.