Awesome Planar Reconstruction:关于平面重建的论文列表,收录了多种单视角、多视角和基于深度学习等方法的重建技术

GitHub:chenzhaiyu/awesome-planar-reconstruction

关于平面重建的论文列表,收录了多种单视角、多视角和基于深度学习等方法的重建技术。涉及到的主题包括平面提取、3D形状检测、深度学习隐函数、多视角重建、网格变形和城市重建等。

Papers

Those with piece-wise planarity are highlighted.

Plane Extraction / Detection

TitleAuthorsVenueYearResources
Efficient RANSAC for Point-Cloud Shape DetectionRuwen Schnabel et al.Computer Graphics Forum2007[PDF] [CODE]
The 3D Hough Transform for Plane Detection in Point Clouds: A Review and a new Accumulator DesignDorit Borrmann et al.3D Research2011[PDF]
Planar Shape Detection at Structural ScalesHao Fang et al.CVPR2018[PDF]
Plane Segmentation Based on the Optimal-vector-field in LiDAR Point CloudsSheng Xu et al.PAMI2020[PDF]
From Planes to Corners: Multi-Purpose Primitive Detection in Unorganized 3D Point CloudsChristiane Sommer et al.ICRA2020[PDF] [CODE]
Supervised Fitting of Geometric Primitives to 3D Point CloudsLingxiao Li et al.CVPR2019[PDF] [CODE]
Highly Parallelizable Plane Extraction for Organized Point Clouds Using Spherical Convex HullsHannes Möls et al.ICRA2020[PDF]
CPFN: Cascaded Primitive Fitting Networks for High-Resolution Point CloudsEric-Tuan Lê et al.ICCV2021[PDF] [CODE]

Single-view Reconstruction

TitleAuthorsVenueYearResources
PlaneNet: Piece-wise Planar Reconstruction from a Single RGB ImageChen Liu et al.CVPR2018[PDF] [CODE]
Recovering 3D Planes from a Single Image via Convolutional Neural NetworksFengting Yang and Zihan ZhouECCV2018[PDF] [CODE]
PlaneRCNN: 3D Plane Detection and Reconstruction from a Single ImageChen Liu et al.CVPR2019[PDF] [CODE]
Single-Image Piece-wise Planar 3D Reconstruction via Associative EmbeddingZehao Yu et al.CVPR2019[PDF] [CODE]
Deep Mesh Reconstruction from Single RGB Images via Topology Modification NetworksJunyi Pan et al.ICCV2019[PDF] [CODE]
Mesh R-CNNGeorgia Gkioxari et al.ICCV2019[PDF] [CODE]
Learning Pairwise Inter-Plane Relations for Piecewise Planar ReconstructionYiming Qian and Yasutaka FurukawaECCV2020[PDF]

Multi-view Reconstruction

TitleAuthorsVenueYearResources
Reconstructing piecewise planar scenes with multi-view regularizationWeijie Xi and Xuejin ChenComputational Visual Media2019[PDF]
PlaneMVS: 3D Plane Reconstruction from Multi-View StereoJiachen Liu et al.CVPR2022[PDF]

Generative / Implicit Field

TitleAuthorsVenueYearResources
Continuous Signed Distance Functions for 3D VisionSimen Haugo et al.3DV2017[PDF]
AtlasNet: A Papier-Mâché Approach to Learning 3D SurfaceThibault Groueix et al.CVPR2018[PDF] [CODE]
DeepSDF: Learning Continuous Signed Distance Functions for Shape RepresentationJeong Joon Park et al.CVPR2019[PDF] [CODE]
Occupancy Networks: Learning 3D Reconstruction in Function SpaceLars Mescheder et al.CVPR2019[PDF] [CODE]
Learning Implicit Fields for Generative Shape ModelingZhiqin Chen and Hao ZhangCVPR2019[PDF] [CODE]
Learning Shape Templates with Structured Implicit FunctionsKyle Genova et al.ICCV2019[PDF]
BSP-Net: Generating Compact Meshes via Binary Space PartitioningZhiqin Chen et al.CVPR2020[PDF] [CODE]
Points2Surf: Learning Implicit Surfaces from Point CloudsPhilipp Erler et al.ECCV2020[PDF] [CODE]
Overfit Neural Networks as a Compact Shape RepresentationThomas Davies et al.arXiv2020[PDF]
Implicit Geometric Regularization for Learning ShapesAmos Gropp et al.ICML2020[PDF] [CODE]
Deep Parametric Shape Predictions using Distance FieldsDmitriy Smirnov et al.CVPR2020[PDF] [CODE]
PolyGen: An Autoregressive Generative Model of 3D MeshesCharlie Nash et al.arXiv2020[PDF]
CvxNet: Learnable Convex DecompositionBoyang Deng et al.arXiv2020[PDF] [CODE]
Local Deep Implicit Functions for 3D ShapeKyle Genova et al.CVPR2020[PDF] [CODE]
Coupling Explicit and Implicit Surface Representations for Generative 3D ModelingOmid Poursaeed et al.ECCV2020[PDF]
CAPRI-Net: Learning Compact CAD Shapes with Adaptive Primitive AssemblyFenggen Yu et al.CVPR2022[PDF]
Neural 3D Scene Reconstruction with the Manhattan-world AssumptionHaoyu Guo et al.CVPR2022[PAGE]
NeuRIS: Neural Reconstruction of Indoor Scenes Using Normal PriorsJiepeng Wang et al.ECCV2022[PAGE]

Kinetic Reconstruction

TitleAuthorsVenueYearResources
KIPPI: KInetic Polygonal Partitioning of ImagesJEAN-PHILIPPE BAUCHET and FLORENT LAFARGECVPR2018[PDF] [CODE]
Approximating shapes in images with low-complexity polygonsMuxingzi Li et al.CVPR2020[PDF] [CODE]
Kinetic Shape ReconstructionJEAN-PHILIPPE BAUCHET and FLORENT LAFARGESIGGRAPH2020[PDF]

Mesh Deformation

TitleAuthorsVenueYearResources
Point2Mesh: A Self-Prior for Deformable MeshesRANA HANOCKA et al.SIGGRAPH2020[PDF] [CODE]
ShapeFlow: Learnable Deformations Among 3D ShapesChiyu Jiang et al.NeurIPS2020[PDF] [CODE]
Neural Mesh Flow: 3D Manifold Mesh Generation via Diffeomorphic FlowsKunal Gupta and Manmohan ChandrakerNeurIPS2020[PDF] [CODE]
Learning Deformable Tetrahedral Meshes for 3D ReconstructionJun Gao et al.NeurIPS2020[PDF] [CODE]

Constructive / Parametric Geometry

TitleAuthorsVenueYearResources
Boolean operations on 3D selective Nef complexes: Data structure, algorithms, optimized implementation and experimentsPeter Hachenberger et al.Computational Geometry2007[PDF] [CODE]
Mesh arrangements for solid geometryQingnan Zhou et al.SIGGRAPH2016[PDF] [CODE]
3D-PRNN: Generating Shape Primitives with Recurrent Neural NetworksChuhang Zou et al.ICCV2017[PDF] [CODE]
CSGNet: Neural Shape Parser for Constructive Solid GeometryGopal Sharma et al.CVPR2018[PDF]
Surface Reconstruction from 3D Line SegmentsPierre-Alain Langlois et al.3DV2019[PDF] [CODE]
ParSeNet: A Parametric Surface Fitting Network for 3D Point CloudsGopal Sharma et al.ECCV2020[PDF] [CODE]
PQ-NET: A Generative Part Seq2Seq Network for 3D Shapes arXiv:1911.10949v3Rundi Wu et al.CVPR2020[PDF] [CODE]
UCSG-Net — Unsupervised Discovering of Constructive Solid Geometry TreeKacper Kania et al.NIPS2020[PDF] [CODE]
State of the Art on Computational Design of Assemblies with Rigid PartsZiqi Wang et al.EUROGRAPHICS2021[PDF]
Reconstruction of Convex Polytope Compositions from 3D Point-cloudsMarkus Friedrich and Pierre-Alain FayolleGRAPP2021[PDF] [CODE]

Triangulation

TitleAuthorsVenueYearResources
PointTriNet: Learned Triangulation of 3D Point SetsNicholas Sharp and Maks OvsjanikovECCV2020[PDF] [CODE]

Wireframe

TitleAuthorsVenueYearResources
PC2WF: 3D Wireframe Reconstruction from Raw Point CloudsYujia Liu et al.ICLR2021[PDF]

Urban Reconstruction

TitleAuthorsVenueYearResources
LOD Generation for Urban ScenesYannick Verdie et al.ACM TOG2015[PDF]
Manhattan-world Urban Reconstruction from Point CloudsMinglei Li et al.ECCV2016[PDF]
Connect-and-Slice: an hybrid approach for reconstructing 3D objectsHao Fang et al.CVPR2020[PDF]
Relation-Constrained Automatic 3D Reconstruction of Buildings in Metropolitan Areas from Photogrammetric Point CloudsYuan Li and Bo WuRemote Sensing2021[PDF]
Combined Rule-Based and Hypothesis-Based Method for Building Model Reconstruction from Photogrammetric Point CloudsLinfu Xie et al.Remote Sensing2021[PDF]
Reconstructing Compact Building Models from Point Clouds Using Deep Implicit FieldsZhaiyu Chen et al.ISPRS2022[PDF] [CODE]
Point2Roof: End-to-end 3D Building Roof Modeling from Airborne LiDAR Point CloudsLi Li et al.ISPRS2022[PDF] [CODE]

License

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.


关于明柳梦少

坚守自己的原则,不随波逐流。