mlsanityData to 7 Oct 2026

DeepMend: Learning Occupancy Functions to Represent Shape for Repair

arXiv 2210.05728 ↗First posted 11 Oct 2022cs.CVcited by 6

Not on the map. Why?

Abstract (arXiv)

We present DeepMend, a novel approach to reconstruct restorations to fractured shapes using learned occupancy functions. Existing shape repair approaches predict low-resolution voxelized restorations, or require symmetries or access to a pre-existing complete oracle. We represent the occupancy of a fractured shape as the conjunction of the occupancy of an underlying complete shape and the fracture surface, which we model as functions of latent codes using neural networks. Given occupancy samples from an input fractured shape, we estimate latent codes using an inference loss augmented with novel penalty terms that avoid empty or voluminous restorations. We use inferred codes to reconstruct the restoration shape. We show results with simulated fractures on synthetic and real-world scanned objects, and with scanned real fractured mugs. Compared to the existing voxel approach and two baseline methods, our work shows state-of-the-art results in accuracy and avoiding restoration artifacts over non-fracture regions of the fractured shape.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 42 of its 63 references that are in the map. First posted on arXiv 11 Oct 2022.

Builds on

42 of 63 references.

References · 42

  1. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  2. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  3. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,682
  4. 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,037
  5. 2018Learning Implicit Fields for Generative Shape Modelingcited by 1,060
  6. 2020Convolutional Occupancy Networkscited by 760
  7. 2016Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modelingcited by 558
  8. 2018AtlasNet: A Papier-Mach'e Approach to Learning 3D Surface Generationcited by 539
  9. 2018PCN: Point Completion Networkcited by 382
  10. 2020Implicit Functions in Feature Space for 3D Shape Reconstruction and Completioncited by 343
  11. 2020Deep Local Shapes: Learning Local SDF Priors for Detailed 3D Reconstructioncited by 321
  12. 2019Learning Shape Templates with Structured Implicit Functionscited by 242
30 more
  1. 2016Shape Completion using 3D-Encoder-Predictor CNNs and Shape Synthesiscited by 219
  2. 2020MetaSDF: Meta-learning Signed Distance Functionscited by 214
  3. 2020Neural-Pull: Learning Signed Distance Functions from Point Clouds by Learning to Pull Space onto Surfacescited by 146
  4. 2019Morphing and Sampling Network for Dense Point Cloud Completioncited by 132
  5. 2016Generative and Discriminative Voxel Modeling with Convolutional Neural Networkscited by 128
  6. 2020Complete & Label: A Domain Adaptation Approach to Semantic Segmentation of LiDAR Point Cloudscited by 116
  7. 2022ShapeFormer: Transformer-based Shape Completion via Sparse Representationcited by 106
  8. 2017ScanComplete: Large-Scale Scene Completion and Semantic Segmentation for 3D Scanscited by 103
  9. 2021Variational Relational Point Completion Networkcited by 103
  10. 2017High-Resolution Shape Completion Using Deep Neural Networks for Global Structure and Local Geometry Inferencecited by 87
  11. 2020Deep Implicit Templates for 3D Shape Representationcited by 86
  12. 2020DualSDF: Semantic Shape Manipulation using a Two-Level Representationcited by 82
  13. 2020PatchNets: Patch-Based Generalizable Deep Implicit 3D Shape Representationscited by 81
  14. 2019SG-NN: Sparse Generative Neural Networks for Self-Supervised Scene Completion of RGB-D Scanscited by 77
  15. 2020SDF-SRN: Learning Signed Distance 3D Object Reconstruction from Static Imagescited by 68
  16. 2017Improved Adversarial Systems for 3D Object Generation and Reconstructioncited by 66
  17. 2020Curriculum DeepSDFcited by 62
  18. 2019RL-GAN-Net: A Reinforcement Learning Agent Controlled GAN Network for Real-Time Point Cloud Shape Completioncited by 58
  19. 2016VConv-DAE: Deep Volumetric Shape Learning Without Object Labelscited by 50
  20. 2022Implicit Autoencoder for Point-Cloud Self-Supervised Representation Learningcited by 46
  21. 2022Neural Shape Mating: Self-Supervised Object Assembly with Adversarial Shape Priorscited by 42
  22. 2018Learning 3D Shape Completion under Weak Supervisioncited by 40
  23. 2020Ladybird: Quasi-Monte Carlo Sampling for Deep Implicit Field Based 3D Reconstruction with Symmetrycited by 24
  24. 2020Dynamic Plane Convolutional Occupancy Networkscited by 24
  25. 2020Deep Optimized Priors for 3D Shape Modeling and Reconstructioncited by 20
  26. 2020Coupling Explicit and Implicit Surface Representations for Generative 3D Modelingcited by 19
  27. 2021Mending Neural Implicit Modeling for 3D Vehicle Reconstruction in the Wildcited by 19
  28. 2020Learning Occupancy Function from Point Clouds for Surface Reconstructioncited by 13
  29. 20173D Reconstruction of Incomplete Archaeological Objects Using a Generative Adversarial Networkcited by 12
  30. 2022Deep Surface Reconstruction from Point Clouds with Visibility Informationcited by 8

Built on by

Cited by year

Citing papers per year: 2023: 2; 2024: 4.

6 papers.

  1. 2023Fantastic Breaks: A Dataset of Paired 3D Scans of Real-World Broken Objects and Their Complete Counterpartscited by 18
  2. 2024Re-assembling the past: The RePAIR dataset and benchmark for real world 2D and 3D puzzle solvingcited by 17
  3. 2024Occupancy-Based Dual Contouringcited by 3
  4. 2024Cultural Heritage 3D Reconstruction with Diffusion Networkscited by 3
  5. 2023Pix2Repair: Implicit Shape Restoration from Imagescited by 0
  6. 20243D Shape Completion with Test-Time Trainingcited by 0