3D分割算法汇总（RGB-D/体素/点云/多模态/实例等任务）

点击下方卡片，关注“自动驾驶之心”公众号

ADAS巨卷干货，即可获取

编辑 | 自动驾驶之心

3D分割是3D空间中的感知任务，直接语义化每个位置的信息，对下游的很多任务有着很大作用，今天为大家盘点下领域常用的算法！

首发于--自动驾驶与AI学习社区：欢迎加入国内首个自动驾驶开发者社区！这里有最全面有效的自动驾驶与AI学习路线（感知/定位/融合）和自动驾驶与AI公司内推机会！

1）基于RGB-D

1. 3D Graph Neural Networks for RGBD Semantic Segmentation

2. Cascaded Feature Network for Semantic Segmentation of RGB-D Images

3. Collaborative Deconvolutional Neural Networks for Joint Depth Estimation and Semantic Segmentation

4. Depth-aware CNN for RGB-D Segmentation

5. Exploiting Depth from Single Monocular Images for Object Detection and Semantic Segmentation

6. FuseNet: Incorporating Depth into Semantic Segmentation via Fusion-based CNN Architecture

7. Incorporating Depth into both CNN and CRF for Indoor Semantic Segmentation

8. Joint Semantic Segmentation and Depth Estimation with Deep Convolutional Networks

9. Learning Common and Speci c Features for RGB-D Semantic Segmentation with Deconvolutional Networks

10. Learning Rich Features from RGB-D Images for Object Detection and Segmentation

11. Locality-Sensitive Deconvolution Networks with Gated Fusion for RGB-D Indoor Semantic Segmentation

12. LSTM-CF: Unifying Context Modeling and Fusion with LSTMs for RGB-D Scene Labeling

13. RGB-D joint modelling with scene geometric information for indoor semantic segmentation

14. RGB-D Scene Labeling with Multimodal Recurrent Neural Networks 6.Semantic segmentation of RGBD images based on deep depth regression

2) 基于体素

1. 3D Semantic Segmentation with Submanifold Sparse Convolutional Networks

2. 3DCNN-DQN-RNN: A Deep Reinforcement Learning Framework for Semantic Parsing of Large-scale 3D Point Clouds

3. Fully-Convolutional Point Networks for Large-Scale Point Clouds

4. OctNet: Learning Deep 3D Representations at High Resolutions

5. ScanComplete: Large-Scale Scene Completion and Semantic Segmentation for 3D Scans

6. SEGCloud: Semantic Segmentation of 3D Point Clouds

7. VV-NET: Voxel VAE Net with Group Convolutions for Point Cloud Segmentation

3）基于点云

基于MLP：

1. 3D Recurrent Neural Networks with Context Fusion for Point Cloud Semantic Segmentation

2. Exploring Spatial Context for 3D Semantic Segmentation of Point Clouds

3. Know What Your Neighbors Do: 3D Semantic Segmentation of Point Clouds

4. PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation

5. PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space

6. PointSIFT: A SIFT-like Network Module for 3D Point Cloud Semantic Segmentation

7. PointWeb: Enhancing Local Neighborhood Features for Point Cloud Processing

基于Point Conv：

1. A-CNN: Annularly Convolutional Neural Networks on Point Clouds

2. Deep Parametric Continuous Convolutional Neural Networks

3. Dilated Point Convolutions: On the Receptive Field Size of Point Convolutions on 3D Point Clouds

4. Flex-Convolution

5. KPConv: Flexible and Deformable Convolution for Point Clouds

6. Monte Carlo Convolution for Learning on Non-Uniformly Sampled Point Clouds

7. PointCNN: Convolution On X-Transformed Points

8. PointConv: Deep Convolutional Networks on 3D Point Clouds

9. Pointwise Convolutional Neural Networks

10. PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation

11. RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds

12. Recurrent Slice Networks for 3D Segmentation of Point Clouds

基于Graph Conv：

1. 3DContextNet: K-d Tree Guided Hierarchical Learning of Point Clouds Using Local and Global Contextual Cues

2. DeepGCNs: Can GCNs Go as Deep as CNNs?

3. Dynamic Graph CNN for Learning on Point Clouds

4. Hierarchical Depthwise Graph Convolutional Neural Network for 3D Semantic Segmentation of Point Clouds

5. Large-scale Point Cloud Semantic Segmentation with Superpoint Graphs

6. Global Context Reasoning for Semantic Segmentation of 3D Point Clouds

7. Octree guided CNN with Spherical Kernels for 3D Point Clouds

8. Point Attention Network for Semantic Segmentation of 3D Point Clouds

9. Point Clouds Learning with Attention-based Graph Convolution Networks

4）基于多模态

1. 3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation

2. A Unified Point-Based Framework for 3D Segmentation

3. Multi-view PointNet for 3D Scene Understanding

4. Point-Voxel CNN for Efficient 3D Deep Learning

5. Sensor Fusion for Joint 3D Object Detection and Semantic Segmentation

5）实例分割

proposal based：

1. 3D-MPA: Multi Proposal Aggregation for 3D Semantic Instance Segmentation

2. 3D-SIS: 3D Semantic Instance Segmentation of RGB-D Scans

3. GSPN: Generative Shape Proposal Network for 3D Instance Segmentation in Point Cloud

4. End-to-end 3D Point Cloud Instance Segmentation without Detection

5. Learning Object Bounding Boxes for 3D Instance Segmentation on Point Clouds

6. SGPN: Similarity Group Proposal Network for 3D Point Cloud Instance Segmentation

proposal free：

1. 3D Bird's-Eye-View Instance Segmentation

2. 3D Graph Embedding Learning with a Structure-aware Loss Function for Point Cloud Semantic Instance Segmentation

3. 3D Instance Segmentation via Multi-Task Metric Learning

4. Associatively Segmenting Instances and Semantics in Point Clouds

5. End-to-end 3D Point Cloud Instance Segmentation without Detection

6. JSIS3D: Joint Semantic-Instance Segmentation of 3D Point Clouds with Multi-Task Pointwise Networks and Multi-Value Conditional Random Fields

7. MASC: Multi-scale Affinity with Sparse Convolution for 3D Instance Segmentation

8. OccuSeg: Occupancy-aware 3D Instance Segmentation

9. PanopticFusion: Online Volumetric Semantic Mapping at the Level of Stuff and Things

10. PointGroup: Dual-Set Point Grouping for 3D Instance Segmentation

① 全网独家视频课程

BEV感知、毫米波雷达视觉融合、多传感器标定、多传感器融合、多模态3D目标检测、点云3D目标检测、目标跟踪、Occupancy、cuda与TensorRT模型部署、协同感知、语义分割、自动驾驶仿真、传感器部署、决策规划、轨迹预测等多个方向学习视频（扫码学习）

视频官网：www.zdjszx.com

② 国内首个自动驾驶学习社区

近2000人的交流社区，涉及30+自动驾驶技术栈学习路线，想要了解更多自动驾驶感知（2D检测、分割、2D/3D车道线、BEV感知、3D目标检测、Occupancy、多传感器融合、多传感器标定、目标跟踪、光流估计）、自动驾驶定位建图（SLAM、高精地图、局部在线地图）、自动驾驶规划控制/轨迹预测等领域技术方案、AI模型部署落地实战、行业动态、岗位发布，欢迎扫描下方二维码，加入自动驾驶之心知识星球，这是一个真正有干货的地方，与领域大佬交流入门、学习、工作、跳槽上的各类难题，日常分享论文+代码+视频，期待交流！

③【自动驾驶之心】技术交流群

自动驾驶之心是首个自动驾驶开发者社区，聚焦目标检测、语义分割、全景分割、实例分割、关键点检测、车道线、目标跟踪、3D目标检测、BEV感知、多模态感知、Occupancy、多传感器融合、transformer、大模型、点云处理、端到端自动驾驶、SLAM、光流估计、深度估计、轨迹预测、高精地图、NeRF、规划控制、模型部署落地、自动驾驶仿真测试、产品经理、硬件配置、AI求职交流等方向。扫码添加汽车人助理微信邀请入群，备注：学校/公司+方向+昵称（快速入群方式）

④【自动驾驶之心】平台矩阵，欢迎联系我们！