mlsanityData to 14 Oct 2026

Reconstructing Surfaces for Sparse Point Clouds with On-Surface Priors

arXiv 2204.10603 ↗First posted 22 Apr 2022cs.CVcited by 60

Not on the map. Why?

Abstract (arXiv)

It is an important task to reconstruct surfaces from 3D point clouds. Current methods are able to reconstruct surfaces by learning Signed Distance Functions (SDFs) from single point clouds without ground truth signed distances or point normals. However, they require the point clouds to be dense, which dramatically limits their performance in real applications. To resolve this issue, we propose to reconstruct highly accurate surfaces from sparse point clouds with an on-surface prior. We train a neural network to learn SDFs via projecting queries onto the surface represented by the sparse point cloud. Our key idea is to infer signed distances by pushing both the query projections to be on the surface and the projection distance to be the minimum. To achieve this, we train a neural network to capture the on-surface prior to determine whether a point is on a sparse point cloud or not, and then leverage it as a differentiable function to learn SDFs from unseen sparse point cloud. Our method can learn SDFs from a single sparse point cloud without ground truth signed distances or point normals. Our numerical evaluation under widely used benchmarks demonstrates that our method achieves state-of-the-art reconstruction accuracy, especially for sparse point clouds.

Key facts. Cited by 60 papers in the map (data through 14 Oct 2026). Builds on 54 of its 70 references that are in the map. First posted on arXiv 22 Apr 2022.

Builds on

54 of 70 references.

References · 54

  1. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  2. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  3. 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,723
  4. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,897
  5. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,689
  6. 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,238
  7. 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,041
  8. 2021Plenoxels: Radiance Fields without Neural Networkscited by 1,613
  9. 2020Convolutional Occupancy Networkscited by 761
  10. 2020Implicit Geometric Regularization for Learning Shapescited by 710
  11. 2021NICE-SLAM: Neural Implicit Scalable Encoding for SLAMcited by 589
  12. 2021Zero-Shot Text-Guided Object Generation with Dream Fieldscited by 513
38 more
  1. 2021Neural Geometric Level of Detail: Real-time Rendering with Implicit 3D Shapescited by 426
  2. 2019SAL: Sign Agnostic Learning of Shapes from Raw Datacited by 390
  3. 2020Local Implicit Grid Representations for 3D Scenescited by 373
  4. 2018ABC: A Big CAD Model Dataset For Geometric Deep Learningcited by 341
  5. 2021Text2Mesh: Text-Driven Neural Stylization for Meshescited by 334
  6. 2020Deep Local Shapes: Learning Local SDF Priors for Detailed 3D Reconstructioncited by 321
  7. 2021Shape As Points: A Differentiable Poisson Solvercited by 187
  8. 2021SnowflakeNet: Point Cloud Completion by Snowflake Point Deconvolution with Skip-Transformercited by 164
  9. 2019SDFDiff: Differentiable Rendering of Signed Distance Fields for 3D Shape Optimizationcited by 159
  10. 2020Neural-Pull: Learning Signed Distance Functions from Point Clouds by Learning to Pull Space onto Surfacescited by 147
  11. 2020Points2Surf: Learning Implicit Surfaces from Point Cloud Patchescited by 122
  12. 2020SALD: Sign Agnostic Learning with Derivativescited by 115
  13. 2018Deep Geometric Prior for Surface Reconstructioncited by 110
  14. 2020Point Cloud Completion by Skip-attention Network with Hierarchical Foldingcited by 106
  15. 2018Point2Sequence: Learning the Shape Representation of 3D Point Clouds with an Attention-based Sequence to Sequence Networkcited by 100
  16. 2020PMP-Net: Point Cloud Completion by Learning Multi-step Point Moving Pathscited by 97
  17. 2022POCO: Point Convolution for Surface Reconstructioncited by 92
  18. 2020PatchNets: Patch-Based Generalizable Deep Implicit 3D Shape Representationscited by 81
  19. 2021DiGS : Divergence guided shape implicit neural representation for unoriented point cloudscited by 77
  20. 2021Deep Implicit Moving Least-Squares Functions for 3D Reconstructioncited by 67
  21. 2019Multi-Angle Point Cloud-VAE: Unsupervised Feature Learning for 3D Point Clouds from Multiple Angles by Joint Self-Reconstruction and Half-to-Half Predictioncited by 66
  22. 2021SA-ConvONet: Sign-Agnostic Optimization of Convolutional Occupancy Networkscited by 59
  23. 2021Cycle4Completion: Unpaired Point Cloud Completion using Cycle Transformation with Missing Region Codingcited by 50
  24. 20223D Shape Reconstruction from 2D Images with Disentangled Attribute Flowcited by 45
  25. 2018View Inter-Prediction GAN: Unsupervised Representation Learning for 3D Shapes by Learning Global Shape Memories to Support Local View Predictionscited by 43
  26. 2022Learning Deep Implicit Functions for 3D Shapes with Dynamic Code Cloudscited by 42
  27. 2020Meshing Point Clouds with Predicted Intrinsic-Extrinsic Ratio Guidancecited by 39
  28. 2020Iso-Points: Optimizing Neural Implicit Surfaces with Hybrid Representationscited by 39
  29. 2019SSRNet: Scalable 3D Surface Reconstruction Networkcited by 35
  30. 2020DRWR: A Differentiable Renderer without Rendering for Unsupervised 3D Structure Learning from Silhouette Imagescited by 35
  31. 2019ShapeCaptioner: Generative Caption Network for 3D Shapes by Learning a Mapping from Parts Detected in Multiple Views to Sentencescited by 30
  32. 2020Sign-Agnostic Implicit Learning of Surface Self-Similarities for Shape Modeling and Reconstruction from Raw Point Cloudscited by 30
  33. 2015SceneNet: Understanding Real World Indoor Scenes With Synthetic Datacited by 29
  34. 2021RetrievalFuse: Neural 3D Scene Reconstruction with a Databasecited by 28
  35. 2018Y^2Seq2Seq: Cross-Modal Representation Learning for 3D Shape and Text by Joint Reconstruction and Prediction of View and Word Sequencescited by 27
  36. 2021Unsupervised Learning of Fine Structure Generation for 3D Point Clouds by 2D Projection Matchingcited by 26
  37. 20193D Shape Completion with Multi-view Consistent Inferencecited by 22
  38. 2021Augmenting Implicit Neural Shape Representations with Explicit Deformation Fieldscited by 19

50 most cited of 54 All 54 (JSON)

Built on by

Cited by year

Citing papers per year: 2022: 7; 2023: 24; 2024: 21; 2025: 8.

60 papers.

Sort:

By area:

  1. 2024You See it, You Got it: Learning 3D Creation on Pose-Free Videos at Scalecited by 65
  2. 2024Neural Signed Distance Function Inference through Splatting 3D Gaussians Pulled on Zero-Level Setcited by 49
  3. 2022CAP-UDF: Learning Unsigned Distance Functions Progressively from Raw Point Clouds with Consistency-Aware Field Optimizationcited by 47
  4. 2022Surface Reconstruction from Point Clouds: A Survey and a Benchmarkcited by 41
  5. 2023Towards Better Gradient Consistency for Neural Signed Distance Functions via Level Set Alignmentcited by 39
  6. 2022NeAF: Learning Neural Angle Fields for Point Normal Estimationcited by 38
  7. 2023Learning a More Continuous Zero Level Set in Unsigned Distance Fields through Level Set Projectioncited by 36
  8. 2023Neural-Singular-Hessian: Implicit Neural Representation of Unoriented Point Clouds by Enforcing Singular Hessiancited by 35
  9. 2022GeoUDF: Surface Reconstruction from 3D Point Clouds via Geometry-guided Distance Representationcited by 29
  10. 2023Unsupervised Inference of Signed Distance Functions from Single Sparse Point Clouds without Learning Priorscited by 28
  11. 2023Learning Signed Distance Functions from Noisy 3D Point Clouds via Noise to Noise Mappingcited by 27
  12. 2024UDiFF: Generating Conditional Unsigned Distance Fields with Optimal Wavelet Diffusioncited by 23
38 more
  1. 2022HSurf-Net: Normal Estimation for 3D Point Clouds by Learning Hyper Surfacescited by 22
  2. 2022Latent Partition Implicit with Surface Codes for 3D Representationcited by 21
  3. 2023Learning Continuous Implicit Field with Local Distance Indicator for Arbitrary-Scale Point Cloud Upsamplingcited by 19
  4. 2024VRP-UDF: Towards Unbiased Learning of Unsigned Distance Functions from Multi-view Images with Volume Rendering Priorscited by 17
  5. 2024Unsupervised Occupancy Learning from Sparse Point Cloudcited by 15
  6. 2024Fast Learning of Signed Distance Functions from Noisy Point Clouds via Noise to Noise Mappingcited by 15
  7. 2025MonoInstance: Enhancing Monocular Priors via Multi-view Instance Alignment for Neural Rendering and Reconstructioncited by 15
  8. 2025GaussianUDF: Inferring Unsigned Distance Functions through 3D Gaussian Splattingcited by 15
  9. 2023NLOS-NeuS: Non-line-of-sight Neural Implicit Surfacecited by 13
  10. 2023Mixing-Denoising Generalizable Occupancy Networkscited by 12
  11. 2025NeRFPrior: Learning Neural Radiance Field as a Prior for Indoor Scene Reconstructioncited by 12
  12. 2023GridPull: Towards Scalability in Learning Implicit Representations from 3D Point Cloudscited by 11
  13. 2023Neural Gradient Learning and Optimization for Oriented Point Normal Estimationcited by 11
  14. 2023Learning Neural Implicit through Volume Rendering with Attentive Depth Fusion Priorscited by 11
  15. 2023Learning Signed Hyper Surfaces for Oriented Point Cloud Normal Estimationcited by 10
  16. 2023NeuralGF: Unsupervised Point Normal Estimation by Learning Neural Gradient Functioncited by 10
  17. 2023Robustifying Generalizable Implicit Shape Networks with a Tunable Non-Parametric Modelcited by 10
  18. 2024Large Point-to-Gaussian Model for Image-to-3D Generationcited by 10
  19. 2022What's the Situation with Intelligent Mesh Generation: A Survey and Perspectivescited by 9
  20. 2023Accurate Differential Operators for Hybrid Neural Fieldscited by 9
  21. 2024A Lightweight UDF Learning Framework for 3D Reconstruction Based on Local Shape Functionscited by 7
  22. 2024Implicit Filtering for Learning Neural Signed Distance Functions from 3D Point Cloudscited by 7
  23. 2024Inferring Neural Signed Distance Functions by Overfitting on Single Noisy Point Clouds through Finetuning Data-Driven based Priorscited by 7
  24. 2024GSurf: Learning Signed Distance Fields from Splatting Opaque Gaussians for High-quality 3D Reconstructioncited by 7
  25. 2025Learning Bijective Surface Parameterization for Inferring Signed Distance Functions from Sparse Point Clouds with Grid Deformationcited by 7
  26. 2023SuperUDF: Self-supervised UDF Estimation for Surface Reconstructioncited by 6
  27. 2024Arbitrary-Scale Point Cloud Upsampling by Voxel-Based Network with Latent Geometric-Consistent Learningcited by 6
  28. 2023Coordinate Quantized Neural Implicit Representations for Multi-view Reconstructioncited by 5
  29. 2023PDF: Point Diffusion Implicit Function for Large-scale Scene Neural Representationcited by 5
  30. 2024Few-Shot Unsupervised Implicit Neural Shape Representation Learning with Spatial Adversariescited by 5
  31. 2024Toward Robust Neural Reconstruction from Sparse Point Setscited by 5
  32. 2025Point2Quad: Generating Quad Meshes from Point Clouds via Face Predictioncited by 5
  33. 2023PG-NeuS: Robust and Efficient Point Guidance for Multi-View Neural Surface Reconstructioncited by 3
  34. 2024PPSURF: Combining Patches and Point Convolutions for Detailed Surface Reconstructioncited by 3
  35. 2024Learning Local Pattern Modularization for Point Cloud Reconstruction from Unseen Classescited by 3
  36. 2023Learning Self-Prior for Mesh Inpainting Using Self-Supervised Graph Convolutional Networkscited by 2
  37. 2023Enhancing Surface Neural Implicits with Curvature-Guided Sampling and Uncertainty-Augmented Representationscited by 2
  38. 2024MergeNet: Explicit Mesh Reconstruction from Sparse Point Clouds via Edge Predictioncited by 2

50 most cited of 60 All 60 (JSON)