ClusterNet解析 ClusterNet: Deep Hierarchical Cluster Network with Rigorously Rotation-Invariant Representation for Point Cloud AnalysisKey points: D维聚类特征构建EdgeConv 多层聚类得到全局特征 ...
PrGANs解析 3D Shape Induction from 2D Views of Multiple ObjectsKey points: 输入201维度的vector(服从U(-1,1)均匀分布) 利用全连接进行shape变化 3D卷积 生成viewpoint，用来将点云投影到平面上 辨别器是通过不同视角的2D图片与投影出的图片进行辨别 dat...
Large-scale Point Cloud Semantic Segmentation with Superpoint Graphs解析 Large-scale Point Cloud Semantic Segmentation with Superpoint GraphsKey points:非监督的超点 超点图 dataset：S3DIS
LPD-Net 解析 LPD-Net: 3D Point Cloud Learning for Large-Scale Place Recognition and Environment AnalysisKey points:hand-crafted传统局部特征提取 通过kNN表示相关 Graph-based Neighborhood Aggregation max pooling 聚合 Robot...
KPConv解析 返回总结列表https://blog.csdn.net/albert199199/article/details/103531486KPConv: Flexible and Deformable Convolution for Point CloudsKey points: 3D conv (kernel): location correlation*Weights L_fit...
PointNet++解析 返回总结列表 https://blog.csdn.net/albert199199/article/details/103531486PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric SpaceKey points: local+global Skip connection...
PointNet解析 返回总结列表 https://blog.csdn.net/albert199199/article/details/103531486PointNet: Deep Learning on Point Sets for 3D Classification and SegmentationKey points: Point-wise 坐标作为channel（N*1的...
3D Deep Learning papers review papers Name Paper Abstract code 2017 PointNet [CVPR] PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation. 多层感知, point-wise, max...
关注
