典型文献
GridNet:efficiently learning deep hierarchical representation for 3D point cloud understanding
文献摘要：
In this paper,we propose a novel and effective ap-proach,namely GridNet,to hierarchically learn deep represen-tation of 3D point clouds.It incorporates the ability of regular holistic description and fast data processing in a single frame-work,which is able to abstract powerful features progressively in an efficient way.Moreover,to capture more accurate internal geometry attributes,anchors are inferred within local neighbor-hoods,in contrast to the fixed or the sampled ones used in ex-isting methods,and the learned features are thus more repre-sentative and discriminative to local point distribution.GridNet delivers very competitive results compared with the state of the art methods in both the object classification and segmentation tasks.
文献关键词：
中图分类号：
作者姓名：
Huiqun WANG;Di HUANG;Yunhong WANG
作者机构：
Laboratory of Intelligent Recognition and Image Processing,School of Computer Science and Engineering,Beihang University,Beijing 100191,China
文献出处：
引用格式：
[1]Huiqun WANG;Di HUANG;Yunhong WANG-.GridNet:efficiently learning deep hierarchical representation for 3D point cloud understanding)[J].计算机科学前沿,2022(01):1-9
A类：
GridNet
B类：
efficiently,learning,deep,representation,point,understanding,In,this,paper,propose,novel,effective,proach,namely,hierarchically,clouds,It,incorporates,ability,regular,holistic,description,fast,data,processing,single,frame,work,which,able,abstract,powerful,features,progressively,way,Moreover,capture,more,accurate,internal,geometry,attributes,anchors,inferred,within,local,neighbor,hoods,contrast,fixed,sampled,ones,used,ex,isting,methods,learned,thus,sentative,discriminative,distribution,delivers,very,competitive,results,compared,state,art,both,object,classification,segmentation,tasks
AB值：
0.656284
相似文献
机标中图分类号，由域田数据科技根据网络公开资料自动分析生成，仅供学习研究参考。
