awesome-point-cloud-analysis-master.md

<style> mark {background-color:#FF9632; font-weight:bold;} </style>

汇总3D点云语义分割论文

参考来源：https://github.com/Yochengliu/awesome-point-cloud-analysis

- Recent papers (from 2017)

此标记代表已经关注

Keywords

dat.: dataset   |   cls.: classification   |   rel.: retrieval   |   seg.: segmentation
det.: detection   |   tra.: tracking   |   pos.: pose   |   dep.: depth
reg.: registration   |   rec.: reconstruction   |   aut.: autonomous driving
oth.: other, including normal-related, correspondence, mapping, matching, alignment, compression, generative model...

Statistics: 🔥 code is available & stars >= 100  |  ⭐ citation >= 50

2017

• [CVPR] PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation. [tensorflow][pytorch] [cls. seg. det.] 🔥 ⭐

• [CVPR] SyncSpecCNN: Synchronized Spectral CNN for 3D Shape Segmentation. [torch] [seg. oth.] ⭐

• [CVPR] ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes. [project][git] [dat. cls. rel. seg. oth.] 🔥 ⭐

• [CVPR] OctNet: Learning Deep 3D Representations at High Resolutions. [torch] [cls. seg. oth.] 🔥 ⭐

• [ICCV] Escape from Cells: Deep Kd-Networks for the Recognition of 3D Point Cloud Models. [pytorch] [cls. rel. seg.] ⭐

• [ICCV] 3DCNN-DQN-RNN: A Deep Reinforcement Learning Framework for Semantic Parsing of Large-scale 3D Point Clouds. [code] [seg.]

• [ICCV] 3D Graph Neural Networks for RGBD Semantic Segmentation. [pytorch] [seg.]

• [NeurIPS] PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space. [tensorflow][pytorch] [cls. seg.] 🔥 ⭐

• [ICRA] Fast segmentation of 3D point clouds: A paradigm on LiDAR data for autonomous vehicle applications. [code] [seg. aut.]

• [ICRA] SegMatch: Segment based place recognition in 3D point clouds. [seg. oth.]

• [3DV] SEGCloud: Semantic Segmentation of 3D Point Clouds. [project] [seg. aut.] ⭐

2018

• [CVPR] SPLATNet: Sparse Lattice Networks for Point Cloud Processing. [caffe] [seg.] 🔥
• [CVPR] Attentional ShapeContextNet for Point Cloud Recognition. [cls. seg.]
• [CVPR] Mining Point Cloud Local Structures by Kernel Correlation and Graph Pooling. [code] [cls. seg.]
• [CVPR] Pointwise Convolutional Neural Networks. [tensorflow] [cls. seg.]
• [CVPR] SO-Net: Self-Organizing Network for Point Cloud Analysis. [pytorch] [cls. seg.] 🔥 ⭐
• [CVPR] Recurrent Slice Networks for 3D Segmentation of Point Clouds. [pytorch] [seg.]
• [CVPR] 3D Semantic Segmentation with Submanifold Sparse Convolutional Networks. [pytorch] [seg.] 🔥
• [CVPR] Deep Parametric Continuous Convolutional Neural Networks. [seg. aut.]
• [CVPR] SGPN: Similarity Group Proposal Network for 3D Point Cloud Instance Segmentation. [tensorflow] [seg.] 🔥
• [CVPR] Large-scale Point Cloud Semantic Segmentation with Superpoint Graphs. [pytorch] [seg.] 🔥
• [CVPR] Density Adaptive Point Set Registration. [code] [reg.]
• [CVPR] A Minimalist Approach to Type-Agnostic Detection of Quadrics in Point Clouds. [seg.]
• [CVPR] PointGrid: A Deep Network for 3D Shape Understanding. [tensorflow] [cls. seg.]
• [CVPR] Tangent Convolutions for Dense Prediction in 3D. [tensorflow] [seg. aut.]
• [ECCV] 3D Recurrent Neural Networks with Context Fusion for Point Cloud Semantic Segmentation. [seg.]
• [ECCV] Local Spectral Graph Convolution for Point Set Feature Learning. [tensorflow] [cls. seg.]
• [ECCV] SpiderCNN: Deep Learning on Point Sets with Parameterized Convolutional Filters. [tensorflow] [cls. seg.]
• [ECCV] Fully-Convolutional Point Networks for Large-Scale Point Clouds. [tensorflow] [seg. oth.]
• [ECCVW] 3DContextNet: K-d Tree Guided Hierarchical Learning of Point Clouds Using Local and Global Contextual Cues. [cls. seg.]
• [NeurIPS] PointCNN: Convolution On X-Transformed Points. [tensorflow][pytorch] [cls. seg.] 🔥
• [TOG] Point Convolutional Neural Networks by Extension Operators. [tensorflow] [cls. seg.]
• [SIGGRAPH Asia] Monte Carlo Convolution for Learning on Non-Uniformly Sampled Point Clouds. [tensorflow] [cls. seg. oth.]
• [SIGGRAPH] Learning local shape descriptors from part correspondences with multi-view convolutional networks. [project] [seg. oth.]
• [MM] RGCNN: Regularized Graph CNN for Point Cloud Segmentation. [tensorflow] [seg.]
• [ICRA] Multi-View 3D Entangled Forest for Semantic Segmentation and Mapping. [seg. oth.]
• [ICRA] SqueezeSeg: Convolutional Neural Nets with Recurrent CRF for Real-Time Road-Object Segmentation from 3D LiDAR Point Cloud. [tensorflow] [seg. aut.]
• [IROS] Extracting Phenotypic Characteristics of Corn Crops Through the Use of Reconstructed 3D Models. [seg. rec.]
• [ACCV] Flex-Convolution (Million-Scale Point-Cloud Learning Beyond Grid-Worlds). [tensorflow] [seg.]
• [arXiv] PointSIFT: A SIFT-like Network Module for 3D Point Cloud Semantic Segmentation. [tensorflow] [seg.] 🔥
• [arXiv] Iterative Transformer Network for 3D Point Cloud. [cls. seg. pos.]
• [arXiv] Multi-column Point-CNN for Sketch Segmentation. [seg.]

---

• [arXiv] Know What Your Neighbors Do: 3D Semantic Segmentation of Point Clouds. [seg.]

2019

• [CVPR] Relation-Shape Convolutional Neural Network for Point Cloud Analysis. [pytorch] [cls. seg. oth.] 🔥
• [CVPR] Spherical Fractal Convolutional Neural Networks for Point Cloud Recognition. [cls. seg.]
• [CVPR] Modeling Point Clouds with Self-Attention and Gumbel Subset Sampling. [cls. seg.]
• [CVPR] A-CNN: Annularly Convolutional Neural Networks on Point Clouds. [tensorflow] [cls. seg.]
• [CVPR] PointConv: Deep Convolutional Networks on 3D Point Clouds. [tensorflow] [cls. seg.] 🔥
• [CVPR] Path-Invariant Map Networks. [tensorflow] [seg. oth.]
• [CVPR] PartNet: A Large-scale Benchmark for Fine-grained and Hierarchical Part-level 3D Object Understanding. [code] [dat. seg.]
• [CVPR] Associatively Segmenting Instances and Semantics in Point Clouds. [tensorflow] [seg.] 🔥
• [CVPR] Octree guided CNN with Spherical Kernels for 3D Point Clouds. [extension] [code] [cls. seg.]
• [CVPR] JSIS3D: Joint Semantic-Instance Segmentation of 3D Point Clouds with Multi-Task Pointwise Networks and Multi-Value Conditional Random Fields. [pytorch] [seg.]
• [CVPR] Point Cloud Oversegmentation with Graph-Structured Deep Metric Learning. [seg.]
• [CVPR] PartNet: A Recursive Part Decomposition Network for Fine-grained and Hierarchical Shape Segmentation. [pytorch] [dat. seg.]
• [CVPR] 4D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networks. [pytorch] [seg.] 🔥
• [CVPR] FilterReg: Robust and Efficient Probabilistic Point-Set Registration using Gaussian Filter and Twist Parameterization. [code] [reg.]
• [CVPR] PointWeb: Enhancing Local Neighborhood Features for Point Cloud Processing. [pytorch] [cls. seg.]
• [CVPR] GSPN: Generative Shape Proposal Network for 3D Instance Segmentation in Point Cloud. [seg.]
• [CVPR] Graph Attention Convolution for Point Cloud Semantic Segmentation. [seg.]
• [CVPR] LP-3DCNN: Unveiling Local Phase in 3D Convolutional Neural Networks. [project] [cls. seg.]
• [CVPR] Structural Relational Reasoning of Point Clouds. [cls. seg.]
• [ICCV] DeepGCNs: Can GCNs Go as Deep as CNNs? [tensorflow] [pytorch] [seg.] 🔥
• [ICCV] KPConv: Flexible and Deformable Convolution for Point Clouds. [tensorflow] [cls. seg.] 🔥
• [ICCV] ShellNet: Efficient Point Cloud Convolutional Neural Networks using Concentric Shells Statistics. [project] [seg.]
• [ICCV] DensePoint: Learning Densely Contextual Representation for Efficient Point Cloud Processing. [pytorch] [cls. seg. oth.]
• [ICCV] Hierarchical Point-Edge Interaction Network for Point Cloud Semantic Segmentation. [seg.]
• [ICCV] Interpolated Convolutional Networks for 3D Point Cloud Understanding. [cls. seg.]
• [ICCV] Dynamic Points Agglomeration for Hierarchical Point Sets Learning. [pytorch] [cls. seg.]
• [ICCV] Unsupervised Multi-Task Feature Learning on Point Clouds. [cls. seg.]
• [ICCV] VV-NET: Voxel VAE Net with Group Convolutions for Point Cloud Segmentation. [tensorflow] [seg.]
• [ICCV] MeteorNet: Deep Learning on Dynamic 3D Point Cloud Sequences. [code] [cls. seg. oth.]
• [ICCV] 3D Instance Segmentation via Multi-Task Metric Learning. [code] [seg.]
• [NeurIPS] Learning Object Bounding Boxes for 3D Instance Segmentation on Point Clouds. [tensorflow] [det. seg.]
• [NeurIPS] Exploiting Local and Global Structure for Point Cloud Semantic Segmentation with Contextual Point Representations. [tensorflow] [seg.]
• [NeurIPS] Point-Voxel CNN for Efficient 3D Deep Learning. [det. seg. aut.]
• [AAAI] Point2Sequence: Learning the Shape Representation of 3D Point Clouds with an Attention-based Sequence to Sequence Network. [tensorflow] [cls. seg.]
• [TOG] Dynamic Graph CNN for Learning on Point Clouds. [tensorflow][pytorch] [cls. seg.] 🔥 ⭐
• [SIGGRAPH Asia] RPM-Net: recurrent prediction of motion and parts from point cloud. [tensorflow] [seg.]
• [SIGGRAPH Asia] StructureNet: Hierarchical Graph Networks for 3D Shape Generation. [seg. oth.]
• [MM] SRINet: Learning Strictly Rotation-Invariant Representations for Point Cloud Classification and Segmentation. [tensorflow] [cls. seg.]
• [MM] Ground-Aware Point Cloud Semantic Segmentation for Autonomous Driving. [code] [seg. aut.]
• [ICRA] SqueezeSegV2: Improved Model Structure and Unsupervised Domain Adaptation for Road-Object Segmentation from a LiDAR Point Cloud. [tensorflow] [seg. aut.]
• [ICRA] Oriented Point Sampling for Plane Detection in Unorganized Point Clouds. [det. seg.]
• [ICRA] PointNetGPD: Detecting Grasp Configurations from Point Sets. [pytorch] [det. seg.]
• [ICRA] Robust 3D Object Classification by Combining Point Pair Features and Graph Convolution. [cls. seg.]
• [ICRA] Hierarchical Depthwise Graph Convolutional Neural Network for 3D Semantic Segmentation of Point Clouds. [seg.]
• [IROS] PASS3D: Precise and Accelerated Semantic Segmentation for 3D Point Cloud. [seg. aut.]
• [IV] End-to-End 3D-PointCloud Semantic Segmentation for Autonomous Driving. [seg.] [aut.]
• [Eurographics Workshop] Generalizing Discrete Convolutions for Unstructured Point Clouds. [pytorch] [cls. seg.]
• [3DV] Rotation Invariant Convolutions for 3D Point Clouds Deep Learning. [project] [cls. seg.]
• [3DV] Effective Rotation-invariant Point CNN with Spherical Harmonics kernels. [tensorflow] [cls. seg. oth.]
• [arxiv] Context Prediction for Unsupervised Deep Learning on Point Clouds. [cls. seg.]
• [arXiv] 3D Graph Embedding Learning with a Structure-aware Loss Function for Point Cloud Semantic Instance Segmentation. [seg.]
• [arXiv] MortonNet: Self-Supervised Learning of Local Features in 3D Point Clouds. [cls. seg.]
• [arXiv] Linked Dynamic Graph CNN: Learning on Point Cloud via Linking Hierarchical Features. [cls. seg.]
• [arXiv] GAPNet: Graph Attention based Point Neural Network for Exploiting Local Feature of Point Cloud. [tensorflow] [cls. seg.]
• [arXiv] Learning Object Bounding Boxes for 3D Instance Segmentation on Point Clouds. [tensorflow] [det. seg.]
• [arXiv] Spatial Transformer for 3D Points. [seg.]
• [arXiv] Point-Voxel CNN for Efficient 3D Deep Learning. [seg. det. aut.]
• [arXiv] SAWNet: A Spatially Aware Deep Neural Network for 3D Point Cloud Processing. [tensorflow] [cls. seg.]
• [arXiv] PyramNet: Point Cloud Pyramid Attention Network and Graph Embedding Module for Classification and Segmentation. [cls. seg.]
• [arXiv] PointAtrousGraph: Deep Hierarchical Encoder-Decoder with Point Atrous Convolution for Unorganized 3D Points. [tensorflow] [cls. seg.]
• [arXiv] Point2SpatialCapsule: Aggregating Features and Spatial Relationships of Local Regions on Point Clouds using Spatial-aware Capsules. [cls. rel. seg.]
• [arXiv] Deformable Filter Convolution for Point Cloud Reasoning. [seg. det. aut.]
• [arXiv] Grid-GCN for Fast and Scalable Point Cloud Learning. [seg. cls.]
• [arXiv] PointPainting: Sequential Fusion for 3D Object Detection. [seg. det.]
• [arXiv] Geometry Sharing Network for 3D Point Cloud Classification and Segmentation. [pytorch] [cls. seg.]
• [arvix] Deep Learning for 3D Point Clouds: A Survey. [code] [cls. det. tra. seg.]
• [arXiv] Point Attention Network for Semantic Segmentation of 3D Point Clouds. [seg.]

2020

• [AAAI] Point2Node: Correlation Learning of Dynamic-Node for Point Cloud Feature Modeling. [seg. cls.]
• [AAAI] PRIN: Pointwise Rotation-Invariant Network. [seg. cls.]
• [CVPR] RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds. [tensorflow] [seg.]
• [WACV] FuseSeg: LiDAR Point Cloud Segmentation Fusing Multi-Modal Data. [seg. aut.]
• [ECCV] PointContrast: Unsupervised Pre-training for 3D Point Cloud Understanding. [cls. seg. det.]

Datasets

• [KITTI] The KITTI Vision Benchmark Suite. [det.]
• [ModelNet] The Princeton ModelNet . [cls.]
• [ShapeNet] A collaborative dataset between researchers at Princeton, Stanford and TTIC. [seg.]
• [PartNet] The PartNet dataset provides fine grained part annotation of objects in ShapeNetCore. [seg.]
• [PartNet] PartNet benchmark from Nanjing University and National University of Defense Technology. [seg.]
• [S3DIS] The Stanford Large-Scale 3D Indoor Spaces Dataset. [seg.]
• [ScanNet] Richly-annotated 3D Reconstructions of Indoor Scenes. [cls. seg.]
• [Stanford 3D] The Stanford 3D Scanning Repository. [reg.]
• [UWA Dataset] . [cls. seg. reg.]
• [Princeton Shape Benchmark] The Princeton Shape Benchmark.
• [SYDNEY URBAN OBJECTS DATASET] This dataset contains a variety of common urban road objects scanned with a Velodyne HDL-64E LIDAR, collected in the CBD of Sydney, Australia. There are 631 individual scans of objects across classes of vehicles, pedestrians, signs and trees. [cls. match.]
• [ASL Datasets Repository(ETH)] This site is dedicated to provide datasets for the Robotics community with the aim to facilitate result evaluations and comparisons. [cls. match. reg. det]
• [Large-Scale Point Cloud Classification Benchmark(ETH)] This benchmark closes the gap and provides a large labelled 3D point cloud data set of natural scenes with over 4 billion points in total. [cls.]
• [Robotic 3D Scan Repository] The Canadian Planetary Emulation Terrain 3D Mapping Dataset is a collection of three-dimensional laser scans gathered at two unique planetary analogue rover test facilities in Canada.
• [Radish] The Robotics Data Set Repository (Radish for short) provides a collection of standard robotics data sets.
• [IQmulus & TerraMobilita Contest] The database contains 3D MLS data from a dense urban environment in Paris (France), composed of 300 million points. The acquisition was made in January 2013. [cls. seg. det.]
• [Oakland 3-D Point Cloud Dataset] This repository contains labeled 3-D point cloud laser data collected from a moving platform in a urban environment.
• [Robotic 3D Scan Repository] This repository provides 3D point clouds from robotic experiments，log files of robot runs and standard 3D data sets for the robotics community.
• [Ford Campus Vision and Lidar Data Set] The dataset is collected by an autonomous ground vehicle testbed, based upon a modified Ford F-250 pickup truck.
• [The Stanford Track Collection] This dataset contains about 14,000 labeled tracks of objects as observed in natural street scenes by a Velodyne HDL-64E S2 LIDAR.
• [PASCAL3D+] Beyond PASCAL: A Benchmark for 3D Object Detection in the Wild. [pos. det.]
• [3D MNIST] The aim of this dataset is to provide a simple way to get started with 3D computer vision problems such as 3D shape recognition. [cls.]
• [WAD] [ApolloScape] The datasets are provided by Baidu Inc. [tra. seg. det.]
• [nuScenes] The nuScenes dataset is a large-scale autonomous driving dataset.
• [PreSIL] Depth information, semantic segmentation (images), point-wise segmentation (point clouds), ground point labels (point clouds), and detailed annotations for all vehicles and people. [paper] [det. aut.]
• [3D Match] Keypoint Matching Benchmark, Geometric Registration Benchmark, RGB-D Reconstruction Datasets. [reg. rec. oth.]
• [BLVD] (a) 3D detection, (b) 4D tracking, (c) 5D interactive event recognition and (d) 5D intention prediction. [ICRA 2019 paper] [det. tra. aut. oth.]
• [PedX] 3D Pose Estimation of Pedestrians, more than 5,000 pairs of high-resolution (12MP) stereo images and LiDAR data along with providing 2D and 3D labels of pedestrians. [ICRA 2019 paper] [pos. aut.]
• [H3D] Full-surround 3D multi-object detection and tracking dataset. [ICRA 2019 paper] [det. tra. aut.]
• [Argoverse BY ARGO AI] Two public datasets (3D Tracking and Motion Forecasting) supported by highly detailed maps to test, experiment, and teach self-driving vehicles how to understand the world around them.[CVPR 2019 paper][tra. aut.]
• [Matterport3D] RGB-D: 10,800 panoramic views from 194,400 RGB-D images. Annotations: surface reconstructions, camera poses, and 2D and 3D semantic segmentations. Keypoint matching, view overlap prediction, normal prediction from color, semantic segmentation, and scene classification. [3DV 2017 paper] [code] [blog]
• [SynthCity] SynthCity is a 367.9M point synthetic full colour Mobile Laser Scanning point cloud. Nine categories. [seg. aut.]
• [Lyft Level 5] Include high quality, human-labelled 3D bounding boxes of traffic agents, an underlying HD spatial semantic map. [det. seg. aut.]
• [SemanticKITTI] Sequential Semantic Segmentation, 28 classes, for autonomous driving. All sequences of KITTI odometry labeled. [ICCV 2019 paper] [seg. oth. aut.]
• [NPM3D] The Paris-Lille-3D has been produced by a Mobile Laser System (MLS) in two different cities in France (Paris and Lille). [seg.]
• [The Waymo Open Dataset] The Waymo Open Dataset is comprised of high resolution sensor data collected by Waymo self-driving cars in a wide variety of conditions. [det.]
• [A*3D: An Autonomous Driving Dataset in Challeging Environments] A*3D: An Autonomous Driving Dataset in Challeging Environments. [det.]
• [PointDA-10 Dataset] Domain Adaptation for point clouds.
• [Oxford Robotcar] The dataset captures many different combinations of weather, traffic and pedestrians. [cls. det. rec.]
• [PandaSet] Public large-scale dataset for autonomous driving provided by Hesai & Scale. It enables researchers to study challenging urban driving situations using the full sensor suit of a real self-driving-car. [det. seg.]
• [3D-FRONT 3D-FUTURE] [Alibaba] 3D-FRONT contains 10,000 houses (or apartments) and ~70,000 rooms with layout information. 3D-FUTURE contains 20,000+ clean and realistic synthetic scenes in 5,000+ diverse rooms which contain 10,000+ unique high quality 3D instances of furniture.