2018cvpr-点云分类、分割等相关论文

Honorable Mention:

SPLATNet: Sparse Lattice Networks for Point Cloud Processing
UMass Amherst & Nvidia
将点视为自高维网格中采样出的稀疏集合的方式来处理点云数据

Spotlight:

PointGrid: A Deep Network for 3D Shape Understanding
University of Missouri – Columbia
Pointnet与网格化的结合，一个网格中可以包含多个点

SGPN: Similarity Group Proposal Network for 3D Point Cloud Instance
Segmentation
University of Southern California
点云数据上的实例分割

Recurrent Slice Networks for 3D Segmentation on Point Clouds
University of Southern California
场景语义分割的新型网络结构，主要将rnn加入当网络结构中

Poster:

Attentional ShapeContextNet for Point Cloud Recognition
University of California, San Diego
点云处理的一种新的形式，区别于体素化或者如同pointnet

Neighbors Do Help: Deeply Exploiting Local Structures of Point Clouds
Clemson University
Arxiv中的题目为：Mining Point Cloud Local Structures by Kernel Correlation and Graph Pooling
切入点类似于pointnet++，针对pointnet中缺失的局部处理，效果优于pointnet++

SO-Net: Self-Organizing Network for Point Cloud Analysis
新加坡国立大学
自组织神经网络在点云数据上的应用

Large-scale Point Cloud Semantic Segmentation with Superpoint Graphs
Universite Paris-Est
通过superpoint graph捕捉点云结构，支持百万级点云数据的处理

PointFusion: Deep Sensor Fusion for 3D Bounding Box Estimation
stanford university

A Network Architecture for Point Cloud Classification via Automatic Depth Images Generation
ethz
将点云数据转化为一系列2d深度图像后，喂入训练好的CNNs进行数据分类

Frustum PointNets for 3D Object Detection from RGB-D Data
Standford university
Pointnet原作者文章，主要关注大规模点云下的目标检测问题

PU-Net: Point Cloud Upsampling Network
CUHK & SIAT
数据驱动的点云数据上采样

Pointwise Convolutional Neural Networks
The University of Tokyo
Pointwise卷积神经网络在点云数据上的应用