计算机工程与应用2024，Vol.60Issue(13)：171-179,9.DOI:10.3778/j.issn.1002-8331.2304-0141
结合区域结构关系和自注意力的边卷积点云分类分割网络
Point Cloud Classification Segmentation Combining Inter-Region Structure Relations and Self-Attention Edge Convolution Network
摘要
Abstract
A new network framework,ISEC-Net(inter-region structure relations and self-attention edge convolution net-work),is proposed to address the problem of insufficient capture of context and relational features within a region in point cloud networks for deep learning.The network consists of two modules:IrConv(inter-region convolution)and SaConv(self-attention convolution).The SaConv module can extract finer edge features,while the IrConv can dynamically inte-grate local structural information into point features and adaptively capture inter-regional relationships.Extensive experi-ments are conducted on the ModelNet40 and ShapeNet datasets for point cloud classification and part segmentation.The results show that on the ModelNet40 dataset,the overall accuracy(OA)of the ISEC-Net model reaches 93.5%,and the average accuracy(mAcc)reaches 90.7% .On the ShapeNet dataset,the average intersection-over-union(mIoU)reaches 86.1%,and the part segmentation accuracy of guitar,headphone,cup and other parts in the single-class intersection-over-union(IoU)experiment is excellent.This demonstrates that compared with traditional dynamic graph convolutional net-works,ISEC-Net can accurately capture the local features and fine structure of point clouds and enhance the aggregation of global features,thus having excellent effectiveness and generalization ability.关键词
边卷积/区域上下文/区域关系/自注意力/深度学习Key words
edge convolution/regional context/regional relationship/self-attention/deep learning分类
计算机与自动化引用本文复制引用
吕志玮,杨家志,周国清,沈露..结合区域结构关系和自注意力的边卷积点云分类分割网络[J].计算机工程与应用,2024,60(13):171-179,9.基金项目
国家自然科学基金(62166012) （62166012）
广西创新驱动发展专项资金项目(桂科AA18118038) （桂科AA18118038）
广西科技基地和人才专项(桂科AD19254002). （桂科AD19254002）
