多尺度测地线距离的点云划分方法  

A multi-scale point cloud partitioning method by geodesic distance-based

作　　者：陈万志 王牧宇 夏羽 汤璇 魏宪 CHEN Wanzhi;WANG Muyu;XIA Yu;TANG Xuan;WEI Xian(School of Software,Liaoning Technical University,Huludao,Liaoning 125105,China;Shanghai Institute of Aerospace System Engineer,Shanghai 201108,China;East China Normal University Software/Hardware Co-Design Engineering Research Center,Ministry of Education,Shanghai 20024l,China)

机构地区：[1]辽宁工程技术大学软件学院,辽宁葫芦岛125105 [2]上海宇航系统工程研究所,上海201108 [3]华东师范大学软硬件协同设计技术与应用教育部工程研究中心,上海200241

出　　处：《测绘科学》2024年第3期58-66,共9页Science of Surveying and Mapping

基　　金：国家重点研发计划项目(2018YFB1403303);辽宁省教育厅高校科研基金项目(2021LJKZ0327)。

摘　　要：针对三维点云目标通常具有复杂的多尺度非欧几里得空间结构,且无序排列、动态性强、难以高精度预测标签,当前的自注意力模块依赖点乘和矩阵转换运算,难以充分捕捉点云对象的多尺度非欧几里得结构等问题,提出了一种新的多尺度自注意力模块——基于测地线距离的多尺度点云划分(GMT),多尺度提取点云数据的非欧几何信息以增强表征能力。GMT在ModelNet40数据集上的整体准确度指标和类别平均准确度指标分别达到93.2%和90.5%;在ScanObjectNN数据集上的整体准确度指标和类别平均准确度指标分别达到82.5%和81.1%,与Point-TnT等其他主流方法相比取得了具有竞争力的结果。实验结果表明,GMT在一定程度上具有较强的非欧几何特征捕捉能力。Point cloud objects are typically composed of complex,unordered,non-Euclidean spatial structures and multi-scale features,often being dynamic and unpredictable.Current self-attention modules rely on dot products and matrix transformations,which make adequately capturing the multi-scale non-Euclidean structure of point cloud objects difficult.This paper proposes Geodesic-based Multi-scale point cloud partitioning Transformer(GMT),a new multi-scale self-attention module to solve the above issue,aiming to extract non-Euclidean geometric information from point clouds at multiple scales to enhance their representation capabilities.GMT achieved an overall accuracy of 93.2%and a mean accuracy of 90.5%on the ModelNet40 dataset.On the ScanObjectNN dataset,it achieved an overall accuracy of 82.5%and a mean accuracy of 81.1%.Compared to other mainstream methods such as Point-TnT,GMT yielded competitive results.Experimental results indicate that GMT possesses a strong capability to capture non-geometric features to some extent.

关 键 词：计算机视觉 点云分类 测地线距离 TRANSFORMER 

分 类 号：P237[天文地球—摄影测量与遥感]