[机器视觉]读点云深度学习论文

计算机视觉顶会(ICCV、CVPR、ECCV)

论文 代码 期刊/年份 作者/学校 备注
PointNet py2.7 CVPR 2017 Charles R. Qi/Stanford University 分类、语义分割,主页
PointNet++ py2.7 NIPS 2017 Charles R. Qi/Stanford University 分类、语义分割,主页
F-PointNet py2.7 CVPR 2018 Charles R. Qi/Stanford University 3D检测,主页
3DMatch Matlab 2016a CVPR 2017 Andy Zeng/Princeton University 主页CVPR2017口头报告4分钟介绍视频
DenseFusion pytorch CVPR 2019 Chen Wang/Stanford University 主页
SGPN tf1.3.0数据下载 CVPR 2018 Weiyue Wang/University of Southern California 点云的实例分割,基于PointNet++,作者主页
ASIS py2.7 Ubuntu1404 CVPR 2019 Xinlong Wang/The University of Adelaide 点云的实例分割、语义分割,作者主页
PointNetLK PyTorch 0.4.0 CVPR 2019 Yasuhiro Aoki/Carnegie Mellon University 点云配准
PointRCNN PyTorch 1.0 CVPR 2019 Shaoshuai Shi/The Chinese University of Hong Kong 二阶段3D检测
The Perfect Match tf CVPR 2019 Zan Gojcic/ETH Zurich 点云配准
PointConv py2.7, tf1.11.0, CUDA9.0, cuDNN7.3, Ubuntu16.04 CVPR 2019 Wenxuan Wu/CORIS Institute 卷积网络
VoteNet Charles R. Qi/Facebook AI Research 3D点云检测
Go-ICPICCV版 c++py ICCV 2013、TPAMI 2016 点云全局配准,主页

机器人顶会(ICRA、IROS、ROBIO):

论文 代码 期刊/年份 作者/学校 备注
Multi-view Self-supervised Deep Learning for 6D Pose Estimation in the Amazon Picking Challenge Matlab 2015b ICRA 2017 Andy Zeng/Princeton University 项目主页

弃看的论文

论文 代码 期刊/年份 作者/学校 备注
PointCNN py3 NeurIPS 2018 Yangyan Li/Shandong University
pointSIFT py3.5 2018 Mingyang Jiang/Shanghai Jiao Tong University 主页
PRIN pytorch 2018 Yang You1/Shanghai Jiao Tong University

【实时点云3D目标检测(PyTorch)】’Complex-YOLO: Real-time 3D Object Detection on Point Clouds pytorch Darknet’ by AI-liu
《3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions》
【用来处理3D点云的Python库】’pyntcloud - pyntcloud is a Python library for working with 3D point clouds.’ by David de la Iglesia Castro
《Iterative Transformer Network for 3D Point Cloud》W Yuan, D Held, C Mertz, M Hebert [CMU] (2018) view

pycuda

精华:
3D Object Detection 3D目标检测
目前物体6自由度位姿估计(6-Dof pose estimation)有哪些主流方法?

重点视频

PointNet、PointNet++B站视频-将门创投 | 斯坦福大学在读博士生祁芮中台:点云上的深度学习及其在三维场景理解中的应用
SGPN分享总结-南加州大学 phd 王薇月:深度学习在点云分割中的应用课程介绍
ASISB站视频-【极市】王鑫龙-CVPR2019联合分割点云中的实例和语义

精华:
点云配准各种方法总结[不定时更新](作者博文有解析PointNet源码的)

以下是配准相关的论文:
CVPR2019论文 The Perfect Match: 3D Point Cloud Matching with Smoothed Densities3DSmoothNet代码(未开放)

CVPR2017论文 3D Point Cloud Registration for Localization using a Deep Neural Network Auto-Encoder代码

ECCV2018论文 3DFeat-Net: Weakly Supervised Local 3D Features for Point Cloud Registration代码

?评价方法 SSRR 2017 3D Registration of Aerial and Ground Robots for Disaster Response: An Evaluation of Features, Descriptors, and Transformation Estimation
Registration with the Point Cloud Library: A Modular Framework for Aligning in 3-DIEEE Robotics & Automation Magazine2015

CVPR 2019 论文汇总(按方向划分,0404 更新中)

3D数据集:
modelnet
ModelNet10.zip、ModelNet40.zip

shapenet