基于拓扑感知和通道注意力的点云分类与分割  

Point Cloud Classification and Segmentation Network Based on Topological Awareness and Channel Attention

作　　者：刘鑫 陈春梅[1,2] 邓豪 刘桂华 袁玲玲[1,2] Liu Xin;Chen Chunmei;Deng Hao;Liu Guihua;Yuan Lingling(School of Information Engineering,Southwest University of Science and Technology,Mianyang 621010,Sichuan,China;Robot Technology Used for Special Environment Key Laboratory of Sichuan Province,Mianyang 621010,Sichuan,China;School of Information Engineering,Mianyang Teachers’College,Mianyang 621000,Sichuan,China;Sichuan HYHZ Tech.Co.,Ltd.,Mianyang 621213,Sichuan,China)

机构地区：[1]西南科技大学信息工程学院,四川绵阳621010 [2]特殊环境机器人技术四川省重点实验室,四川绵阳621010 [3]绵阳师范学院信息工程学院,四川绵阳621000 [4]四川华越汇智科技有限责任公司,四川绵阳621213

出　　处：《激光与光电子学进展》2024年第24期171-180,共10页Laser & Optoelectronics Progress

基　　金：中央军委装备发展部(23ZG8102);校博士基金(20zx7123)。

摘　　要：针对基于深度学习点云处理方法因浅层几何信息提取不充分、高层级语义特征损失导致分类分割效果不佳的问题,提出一种基于拓扑感知和通道注意力的点云分类分割网络,从信息扩增及特征增强两方面提升点云分类分割的准确性。首先,针对点云数据无序性导致的浅层信息特征表达弱的问题,利用局部最小三角策略构建了点云数据与邻域点的拓扑关系;然后,构建了残差多层感知器模块以提取更为精细的的局部几何特征;最后,采用混合池化对局部信息进行特征聚合,并结合通道亲和注意力机制减少网络中的特征损失。分别在ModelNet40、ScanObjectNN、ShapeNet Part以及S3DIS数据集上进行了大量对比实验,分类任务上总体分类精度分别达到了93.6%、85.6%,分割任务上平均交并比分别达到了85.8%、63.7%,实验结果证明了所提方法在点云分类分割方面性能优异。The unsatisfied effects of point cloud classification and segmentation caused by insufficient geometric feature extraction and high-level semantic feature loss in deep learning-based point cloud processing methods are problematic.Hence,this study proposes a point cloud classification and segmentation network based on topological awareness and channel attention to address these issues.The proposed method improves the precision of point cloud classification and segmentation from information amplification and feature enhancement.First,for the weak expression of low-level geometric features caused by the point cloud data being disordered,the topological relationships between the point cloud data and their neighborhood points are constructed using the local minimum triangulation strategy.Then,the residual multilayer perceptron module is applied to extract more granular local geometric information.Finally,the mixed pooling strategy is exploited for the feature aggregation of local information,and the channel affinity attention mechanism is used to reduce the feature loss in the network.Extensive comparative experiments are conducted on the ModelNet40 dataset,ScanObjectNN dataset,ShapeNet Part dataset,and S3DIS dataset.Using the proposed method,overall accuracy values of 93.6%and 85.6%are achieved in a classification task.Moreover,average crossover ratio values of 85.8%and 63.7%of are achieved in a segmentation task.Therefore,the experimental results demonstrate that the proposed method displays state-of-the-art performance in both point cloud classification and segmentation tasks.

关 键 词：机器视觉 拓扑感知 注意力机制 点云分类与分割 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]