Skip to content

bluestyle97/awesome-3d-reconstruction-papers

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 

Repository files navigation

Awesome 3D Reconstruction Papers

Awesome

A collection of 3D reconstruction papers in the deep learning era. Feel free to contribute :)

Table of Contents

Object-level

Single-view

Paper Representation Publisher Project/Code
A Point Set Generation Network for 3D Object Reconstruction from a Single Image Point Cloud CVPR 2017 Code
SurfNet: Generating 3D Shape Surfaces Using Deep Residual Networks Mesh CVPR 2017 Code
Multi-View Supervision for Single-View Reconstruction via Differentiable Ray Consistency Voxel CVPR 2017 Project
OctNet: Learning Deep 3D Representations at High Resolutions Voxel CVPR 2017 Code
Rethinking Reprojection: Closing the Loop for Pose-Aware Shape Reconstruction From a Single Image Voxel ICCV 2017 /
MarrNet: 3D Shape Reconstruction via 2.5D Sketches Voxel NIPS 2017 Project
Hierarchical Surface Prediction for 3D Object Reconstruction Voxel 3DV 2017 /
Image2Mesh: A Learning Framework for Single Image 3D Reconstruction Mesh ACCV 2018 Code
Learning Efficient Point CloudGeneration for Dense 3D Object Reconstruction Point Cloud AAAI 2018 Project
A Papier-Mâché Approach to Learning 3D Surface Generation Mesh CVPR 2018 Project
Pixels, voxels, and views: A study of shape representations for single view 3D object shape prediction Generic CVPR 2018 Project
Im2Struct: Recovering 3D Shape Structure From a Single RGB Image Parts CVPR 2018 Code
Matryoshka Networks: Predicting 3D Geometry via Nested Shape Layers Voxel CVPR 2018 Code
Multi-View Consistency as Supervisory Signal for Learning Shape and Pose Prediction Voxel CVPR 2018 Project
Efficient Dense Point Cloud Object Reconstruction using Deformation Vector Fields Point Cloud ECCV 2018 /
GAL: Geometric Adversarial Loss for Single-View 3D-Object Reconstruction Point Cloud ECCV 2018 /
Learning Category-Specific Mesh Reconstruction from Image Collections Mesh ECCV 2018 Project
Learning Shape Priors for Single-View 3D Completion and Reconstruction Voxel ECCV 2018 /
Learning Single-View 3D Reconstruction with Limited Pose Supervision Voxel ECCV 2018 Code
Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images Mesh ECCV 2018 Code
Residual MeshNet: Learning to Deform Meshes for Single-View 3D Reconstruction Mesh 3DV 2018 /
Learning to Reconstruct Shapes from Unseen Classes Generic NIPS 2018 /
Multi-View Silhouette and Depth Decomposition for High Resolution 3D Object Representation Voxel NIPS 2018 Code
MVPNet: Multi-View Point Regression Networks for 3D Object Reconstruction from A Single Image Point Cloud AAAI 2019 /
Deep Single-View 3D Object Reconstruction with Visual Hull Embedding Voxel AAAI 2019 Code
Occupancy Networks: Learning 3D Reconstruction in Function Space Implicit CVPR 2019 Code
Learning Implicit Fields for Generative Shape Modeling Implicit CVPR 2019 Project
A Skeleton-Bridged Deep Learning Approach for Generating Meshes of Complex Topologies From Single RGB Images Mesh CVPR 2019 Code
What Do Single-view 3D Reconstruction Networks Learn? Generic CVPR 2019 Code
Deep Level Sets: Implicit Surface Representations for 3D Shape Inference Implicit arXiv 2019 /
Deep Mesh Reconstruction From Single RGB Images via Topology Modification Networks Mesh ICCV 2019 Code
Deep Meta Functionals for Shape Representation Implicit ICCV 2019 Code
GraphX-Convolution for Point Cloud Deformation in 2D-to-3D Conversion Point Cloud ICCV 2019 Code
Pix2Vox: Context-Aware 3D Reconstruction From Single and Multi-View Images Voxel ICCV 2019 Code
Domain-Adaptive Single-View 3D Reconstruction Voxel ICCV 2019 Code
Few-Shot Generalization for Single-Image 3D Reconstruction via Priors Voxel ICCV 2019 Code
DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstruction Implicit NIPS 2019 Code
Front2Back: Single View 3D Shape Reconstruction via Front to Back Prediction Mesh CVPR 2020 Code
BSP-Net: Generating Compact Meshes via Binary Space Partitioning Mesh CVPR 2020 Project
Height and Uprightness Invariance for 3D Prediction From a Single View Point Cloud CVPR 2020 Code
Implicit Functions in Feature Space for 3D Shape Reconstruction and Completion Implicit CVPR 2020 Project
Unsupervised Learning of Probably Symmetric Deformable 3D Objects From Images in the Wild Mesh CVPR 2020 Project
CvxNet: Learnable Convex Decomposition Primitive CVPR 2020 Project
Deep Local Shapes: Learning Local SDF Priors for Detailed 3D Reconstruction Implicit ECCV 2020 /
Few-Shot Single-View 3-D Object Reconstruction with Compositional Priors Voxel ECCV 2020 Code
GSIR: Generalizable 3D Shape Interpretation and Reconstruction Voxel ECCV 2020 /
DR-KFS: A Differentiable Visual Similarity Metric for 3D Shape Reconstruction Mesh ECCV 2020 /
Self-supervised Single-view 3D Reconstruction via Semantic Consistency Mesh ECCV 2020 Project
Shape and Viewpoint without Keypoints Mesh ECCV 2020 Project
Ladybird: Quasi-Monte Carlo Sampling for Deep Implicit Field Based 3D Reconstruction with Symmetry Mesh ECCV 2020 Code
Learning to Detect 3D Reflection Symmetry for Single-View Reconstruction / arXiv 2020 Code
3D Reconstruction of Novel Object Shapes from Single Images Implicit arXiv 2020 Project
3D Shape Reconstruction from Free-Hand Sketches Point Cloud arXiv 2020 Code
Implicit Mesh Reconstruction from Unannotated Image Collections Mesh arXiv 2020 Project
SkeletonNet: A Topology-Preserving Solution for Learning Mesh Reconstruction of Object Surfaces from RGB Images Mesh arXiv 2020 Code
Learning Deformable Tetrahedral Meshes for 3D Reconstruction Mesh NIPS 2020 Project
SDF-SRN: Learning Signed Distance 3D Object Reconstruction from Static Images Implicit NIPS 2020 Project
UCLID-Net: Single View Reconstruction in Object Space Mesh NIPS 2020 Code
Pix2Vox++: Multi-scale Context-aware 3D Object Reconstruction from Single and Multiple Images Voxel IJCV 2020 Code
D2IM-Net: Learning Detail Disentangled Implicit Fields From Single Images Implicit CVPR 2021 Code
Fostering Generalization in Single-view 3D Reconstruction by Learning a Hierarchy of Local and Global Shape Priors Implicit CVPR 2021 Code
Single-View 3D Object Reconstruction From Shape Priors in Memory Voxel CVPR 2021 Project
Look, Cast and Mold: Learning 3D Shape Manifold from Single-view Synthetic Data Voxel arXiv 2021 /
Neural Implicit 3D Shapes from Single Images with Spatial Patterns Implicit arXiv 2021 Code
Implicit Surface Representations as Layers in Neural Networks Implicit ICCV 2021 /
Ray-ONet: Efficient 3D Reconstruction From A Single RGB Image Implicit BMVC 2021 Project
Learning Anchored Unsigned Distance Functions with Gradient Direction Alignment for Single-view Garment Reconstruction Implicit ICCV 2021 Code
Geometric Granularity Aware Pixel-to-Mesh Mesh ICCV 2021 /
Sketch2Mesh: Reconstructing and Editing 3D Shapes from Sketches Mesh ICCV 2021 Code
3DIAS: 3D Shape Reconstruction With Implicit Algebraic Surfaces Primitive ICCV 2021 Project
A Dataset-Dispersion Perspective on Reconstruction Versus Recognition in Single-View 3D Reconstruction Networks Point Cloud 3DV 2021 Code
AutoSDF: Shape Priors for 3D Completion, Reconstruction and Generation Implicit CVPR 2022 Project
3D Shape Reconstruction from 2D Images with Disentangled Attribute Flow Point Cloud CVPR 2022 Code
Pre-train, Self-train, Distill: A simple recipe for Supersizing 3D Reconstruction Implicit CVPR 2022 Project
Neural Template: Topology-aware Reconstruction and Disentangled Generation of 3D Meshes Hybrid CVPR 2022 Code
Training Data Generating Networks: Shape Reconstruction via Bi-level Optimization Implicit ICLR 2022 /
Structural Causal 3D Reconstruction Hybrid ECCV 2022 /
Few-shot Single-view 3D Reconstruction with Memory Prior Contrastive Network Voxel ECCV 2022 /

Multi-view

Paper Representation Publisher Project/Code
3D-R2N2: A Unified Approach for Single and Multi-view 3D Object Reconstruction Voxel ECCV 2016 Code
3D Shape Induction from 2D Views of Multiple Objects Voxel 3DV 2017 Code
Learning Efficient Point Cloud Generation for Dense 3D Object Reconstruction Point Cloud AAAI 2018 Project
Conditional Single-view Shape Generation for Multi-view Stereo Reconstruction Point Cloud CVPR 2019 Code
Pixel2Mesh++: Multi-View 3D Mesh Generation via Deformation Mesh ICCV 2019 Project
Multiview Aggregation for Learning Category-Specific Shape Reconstruction Point Cloud NIPS 2019 Code
Pix2Surf: Learning Parametric 3D Surface Models of Objects from Images Patches ECCV 2020 Project
Multi-view 3D Reconstruction with Transformers Voxel ICCV 2021 /
3D-C2FT: Coarse-to-fine Transformer for Multi-view 3D Reconstruction Voxel arXiv 2022 /
FvOR: Robust Joint Shape and Pose Optimization for Few-view Object Reconstruction Implicit CVPR 2022 Code

Unsupervised

Paper Representation Publisher Project/Code
Perspective Transformer Nets: Learning Single-View 3D Object Reconstruction without 3D Supervision Voxel NIPS 2016 Code
Multi-view Supervision for Single-View Reconstruction via Differentiable Ray Consistency Voxel CVPR 2017 Project
Rethinking Reprojection: Closing the Loop for Pose-Aware Shape Reconstruction from a Single Image Voxel ICCV 2017 /
Learning Category-Specific Mesh Reconstruction from Image Collections Mesh ECCV 2018 Project
Learning Single-View 3D Reconstruction with Limited Pose Supervision Voxel ECCV 2018 Code
Multi-View Consistency as Supervisory Signal for Learning Shape and Pose Prediction Voxel CVPR 2018 Project
Learning View Priors for Single-view 3D Reconstruction Mesh CVPR 2019 Code
Escaping Plato's Cave: 3D Shape From Adversarial Rendering Voxel ICCV 2019 Project
Learning to Infer Implicit Surfaces without 3D Supervision Implicit NIPS 2019 /
Learning to Predict 3D Objects with an Interpolation-based Differentiable Renderer Mesh NIPS 2019 Project
Unsupervised Learning of Probably Symmetric Deformable 3D Objects from Images in the Wild Mesh CVPR 2020 Project
Leveraging 2D Data to Learn Textured 3D Mesh Generation Mesh CVPR 2020 Code
Implicit Mesh Reconstruction from Unannotated Image Collections Mesh arXiv 2020 Project
Shape and Viewpoint without Keypoints Mesh ECCV 2020 Project
Self-supervised Single-view 3D Reconstruction via Semantic Consistency Mesh ECCV 2020 Project
SDF-SRN: Learning Signed Distance 3D Object Reconstruction from Static Images Implicit NIPS 2020 Project
Shelf-Supervised Mesh Prediction in the Wild Mesh CVPR 2021 Project
Fully Understanding Generic Objects: Modeling, Segmentation, and Reconstruction Implicit CVPR 2021 Project
Self-Supervised 3D Mesh Reconstruction from Single Images Mesh CVPR 2021 Code
Do 2D GANs Know 3D Shape? Unsupervised 3D shape reconstruction from 2D Image GANs Mesh ICLR 2021 Project
Discovering 3D Parts from Image Collections Mesh ICCV 2021 Project
Learning Canonical 3D Object Representation for Fine-Grained Recognition Mesh ICCV 2021 /
Toward Realistic Single-View 3D Object Reconstruction with Unsupervised Learning from Multiple Images Mesh ICCV 2021 Code
Learning Generative Models of Textured 3D Meshes from Real-World Images Mesh ICCV 2021 Code
To The Point: Correspondence-driven monocular 3D category reconstruction Mesh NIPS 2021 Project
Share With Thy Neighbors: Single-View Reconstruction by Cross-Instance Consistency Mesh arXiv 2022 Project
Topologically-Aware Deformation Fields for Single-View 3D Reconstruction Implicit CVPR 2022 Project
2D GANs Meet Unsupervised Single-View 3D Reconstruction Implicit ECCV 2022 Project

Scene-level

Single-view

Paper Representation Publisher Project/Code
IM2CAD CAD CVPR 2017 Code
3D-RCNN: Instance-level 3D Object Reconstruction via Render-and-Compare Priors CVPR 2018 Project
Factoring Shape, Pose, and Layout from the 2D Image of a 3D Scene Voxel CVPR 2018 Project
Holistic 3D Scene Parsing and Reconstruction from a Single RGB Image CAD ECCV 2018 Project
Mesh R-CNN Mesh ICCV 2019 Code
3D Scene Reconstruction With Multi-Layer Depth and Epipolar Transformers Mesh ICCV 2019 Project
3D-RelNet: Joint Object and Relational Network for 3D Prediction Voxel ICCV 2019 Project
Total3DUnderstanding: Joint Layout, Object Pose and Mesh Reconstruction for Indoor Scenes from a Single Image Mesh CVPR 2020 Project
3D Scene Reconstruction from a Single Viewport Voxel ECCV 2020 Code
CoReNet: Coherent 3D scene reconstruction from a single RGB image Voxel+Implicit ECCV 2020 Code
Image-to-Voxel Model Translation for 3D Scene Reconstruction and Segmentation Voxel ECCV 2020 Code
Holistic 3D Scene Understanding from a Single Image with Implicit Representation Implicit CVPR 2021 Project
From Points to Multi-Object 3D Reconstruction Implicit CVPR 2021 Project
Learning to Recover 3D Scene Shape from a Single Image Point Cloud CVPR 2021 Code
Patch2CAD: Patchwise Embedding Learning for In-the-Wild Shape Retrieval from a Single Image CAD ICCV 2021 /
Panoptic 3D Scene Reconstruction From a Single RGB Image Voxel NIPS 2021 Project
Voxel-based 3D Detection and Reconstruction of Multiple Objects from a Single Image Implicit NIPS 2021 Project
Towards High-Fidelity Single-view Holistic Reconstruction of Indoor Scenes Implicit ECCV 2022 Code

Multi-view

Paper Representation Publisher Project/Code
MARMVS: Matching Ambiguity Reduced Multiple View Stereo for Efficient Large Scale Scene Reconstruction Point Cloud CVPR 2020 /
FroDO: From Detections to 3D Objects Implicit CVPR 2020 /
Associative3D: Volumetric Reconstruction from Sparse Views Voxel ECCV 2020 Project
Atlas: End-to-End 3D Scene Reconstruction from Posed Images Mesh ECCV 2020 Project
NeuralRecon: Real-Time Coherent 3D Reconstruction from Monocular Video Mesh CVPR 2021 Project
TransformerFusion: Monocular RGB Scene Reconstruction using Transformers Implicit NIPS 2021 Project
Learning 3D Object Shape and Layout without 3D Supervision Mesh CVPR 2022 Project

Neural Surface

Multi-view

Paper Representation Publisher Project/Code
SDFDiff: Differentiable Rendering of Signed Distance Fields for 3D Shape Implicit CVPR 2020 Code
Differentiable Volumetric Rendering: Learning Implicit 3D Representations without 3D Supervision Implicit CVPR 2020 Code
Multiview Neural Surface Reconstruction by Disentangling Geometry and Appearance Implicit NIPS 2020 Project
Unsupervised Learning of 3D Object Categories from Videos in the Wild Implicit CVPR 2021 Project
Neural Lumigraph Rendering Implicit CVPR 2021 Project
Iso-Points: Optimizing Neural Implicit Surfaces With Hybrid Representations Implicit CVPR 2021 Project
Learning Signed Distance Field for Multi-view Surface Reconstruction Implicit ICCV 2021 Code
UNISURF: Unifying Neural Implicit Surfaces and Radiance Fields for Multi-View Reconstruction Implicit ICCV 2021 Project
NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction Implicit NIPS 2021 Project
NeRS: Neural Reflectance Surfaces for Sparse-View 3D Reconstruction in the Wild Implicit NIPS 2021 Project
Volume Rendering of Neural Implicit Surfaces Implicit ICCV 2021 Unofficial Code
NeuralWarp: Improving neural implicit surfaces geometry with patch warping Implicit CVPR 2022 Project
Neural 3D Scene Reconstruction with the Manhattan-world Assumption Implicit CVPR 2022 Project
GenDR: A Generalized Differentiable Renderer Mesh CVPR 2022 Code
NeRFusion: Fusing Radiance Fields for Large-Scale Scene Reconstruction Implicit CVPR 2022 Project
Critical Regularizations for Neural Surface Reconstruction in the Wild Implicit CVPR 2022 /
Multi-View Mesh Reconstruction with Neural Deferred Shading Mesh CVPR 2022 Project
Differentiable Stereopsis: Meshes From Multiple Views Using Differentiable Rendering Mesh CVPR 2022 Code
SparseNeuS: Fast Generalizable Neural Surface Reconstruction from Sparse views Implicit ECCV 2022 Project
Neural 3D Reconstruction in the Wild Implicit SIGGRAPH 2022 Project
Geo-Neus: Geometry-Consistent Neural Implicit Surfaces Learning for Multi-view Reconstruction Implicit arXiv 2022 Code
MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstruction Implicit arXiv 2022 Project
SNeS: Learning Probably Symmetric Neural Surfaces from Incomplete Data Implicit arXiv 2022 /
NeuRIS: Neural Reconstruction of Indoor Scenes Using Normal Priors Implicit arXiv 2022 Project
Differentiable Signed Distance Function Rendering Implicit SIGGRAPH 2022 Project
Differentiable Rendering of Neural SDFs through Reparameterization Implicit arXiv 2022 /

Point-cloud

Paper Representation Publisher Project/Code
Deep Geometric Prior for Surface Reconstruction Patches CVPR 2019 Code
Scan2Mesh: From Unstructured Range Scans to 3D Meshes Mesh CVPR 2019 /
Meshlet Priors for 3D Mesh Reconstruction Mesh CVPR 2020 /
SSRNet: Scalable 3D Surface Reconstruction Network Implicit CVPR 2020 /
SAL: Sign Agnostic Learning of Shapes from Raw Data Implicit CVPR 2020 Code
Implicit Geometric Regularization for Learning Shapes Implicit ICML 2020 Code
Meshing Point Clouds with Predicted Intrinsic-Extrinsic Ratio Guidance Mesh ECCV 2020 Code
PointTriNet: Learned Triangulation of 3D Point Sets Mesh ECCV 2020 Code
Points2Surf: Learning Implicit Surfaces from Point Cloud Patches Implicit ECCV 2020 Code
Convolutional Occupancy Networks Implicit ECCV 2020 Code
Implicit Neural Representations with Periodic Activation Functions Implicit NIPS 2020 Project
Neural Unsigned Distance Fields for Implicit Function Learning Implicit NIPS 2020 Code
Differentiable Surface Triangulation Mesh TOG 2021 Code
SALD: Sign Agnostic Learning with Derivatives Implicit ICLR 2021 Code
Deep Implicit Moving Least-Squares Functions for 3D Reconstruction Implicit CVPR 2021 Code
Sign-Agnostic Implicit Learning of Surface Self-Similarities for Shape Modeling and Reconstruction from Raw Point Clouds Implicit CVPR 2021 /
Learning Delaunay Surface Elements for Mesh Reconstruction Mesh CVPR 2021 Code
Neural Splines: Fitting 3D Surfaces with Infinitely-Wide Neural Networks Implicit CVPR 2021 Code
Neural-Pull: Learning Signed Distance Functions from Point Clouds by Learning to Pull Space onto Surfaces Implicit ICML 2021 Code
Phase Transitions, Distance Functions, and Implicit Neural Representations Implicit ICML 2021 /
Vis2Mesh: Efficient Mesh Reconstruction from Unstructured Point Clouds of Large Scenes with Learned Virtual View Visibility Mesh ICCV 2021 Code
Deep Hybrid Self-Prior for Full 3D Mesh Generation Mesh ICCV 2021 /
Adaptive Surface Reconstruction with Multiscale Convolutional Kernels Mesh ICCV 2021 Code
SA-ConvONet: Sign-Agnostic Optimization of Convolutional Occupancy Networks Implicit ICCV 2021 Code
Deep Implicit Surface Point Prediction Networks Implicit ICCV 2021 Project
Shape As Points: A Differentiable Poisson Solver Mesh NIPS 2021 Project
AIR-Nets: An Attention-Based Framework for Locally Conditioned Implicit Representations Implicit 3DV 2021 Code
Scalable Surface Reconstruction with Delaunay-Graph Neural Networks Mesh arXiv 2021 /
Neural-IMLS: Learning Implicit Moving Least-Squares for Surface Reconstruction from Unoriented Point clouds Implicit arXiv 2021 Project
Neural Fields as Learnable Kernels for 3D Reconstruction Implicit arXiv 2021 Project
POCO: Point Convolution for Surface Reconstruction Implicit CVPR 2022 Code
GIFS: Neural Implicit Function for General Shape Representation Implicit CVPR 2022 Project
Reconstructing Surfaces for Sparse Point Clouds with On-Surface Priors Implicit CVPR 2022 Code
Surface Reconstruction from Point Clouds by Learning Predictive Context Priors Implicit CVPR 2022 Code
DiGS: Divergence Guided Shape Implicit Neural Representation for Unoriented Point Clouds Implicit CVPR 2022 Project
Dual Octree Graph Networks for Learning Adaptive Volumetric Shape Representations Implicit SIGGRAPH 2022 Code
Deep Point Cloud Simplification for High-quality Surface Reconstruction Implicit arXiv 2022 /
GenSDF: Two-Stage Learning of Generalizable Signed Distance Functions Implicit arXiv 2022 Code
RangeUDF: Semantic Surface Reconstruction from 3D Point Clouds Implicit arXiv 2022 Code

RGB-D

Paper Representation Publisher Project/Code
Neural RGB-D Surface Reconstruction Implicit CVPR 2022 Project
BNV-Fusion: Dense 3D Reconstruction Using Bi-Level Neural Volume Fusion Implicit CVPR 2022 Code
NICE-SLAM: Neural Implicit Scalable Encoding for SLAM Implicit CVPR 2022 Project
GO-Surf: Neural Feature Grid Optimization for Fast, High-Fidelity RGB-D Surface Reconstruction Implicit 3DV 2022 Project

Survey

Paper Publisher
Image-based 3D Object Reconstruction: State-of-the-Art and Trends in the Deep Learning Era TPAMI 2019
Surface Reconstruction from Point Clouds: A Survey and a Benchmark arXiv 2022

About

A collection of 3D reconstruction papers in the deep learning era.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •