mlsanityData to 7 Oct 2026

SweepNet: Unsupervised Learning Shape Abstraction via Neural Sweepers

arXiv 2407.06305 ↗First posted 8 Jul 2024cs.CV cs.GRcited by 5

Not on the map. Why?

Abstract (arXiv)

Shape abstraction is an important task for simplifying complex geometric structures while retaining essential features. Sweep surfaces, commonly found in human-made objects, aid in this process by effectively capturing and representing object geometry, thereby facilitating abstraction. In this paper, we introduce papername, a novel approach to shape abstraction through sweep surfaces. We propose an effective parameterization for sweep surfaces, utilizing superellipses for profile representation and B-spline curves for the axis. This compact representation, requiring as few as 14 float numbers, facilitates intuitive and interactive editing while preserving shape details effectively. Additionally, by introducing a differentiable neural sweeper and an encoder-decoder architecture, we demonstrate the ability to predict sweep surface representations without supervision. We show the superiority of our model through several quantitative and qualitative experiments throughout the paper. Our code is available at https://mingrui-zhao.github.io/SweepNet/

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 51 of its 82 references that are in the map. First posted on arXiv 8 Jul 2024.

Builds on

51 of 82 references.

References · 51

  1. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  2. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,682
  3. 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,037
  4. 2018Learning Implicit Fields for Generative Shape Modelingcited by 1,060
  5. 2019PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitizationcited by 818
  6. 2020Convolutional Occupancy Networkscited by 760
  7. 2020Local Implicit Grid Representations for 3D Scenescited by 372
  8. 2020Implicit Functions in Feature Space for 3D Shape Reconstruction and Completioncited by 343
  9. 2019DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstructioncited by 332
  10. 2020Deep Local Shapes: Learning Local SDF Priors for Detailed 3D Reconstructioncited by 321
  11. 2021DeepCAD: A Deep Generative Network for Computer-Aided Design Modelscited by 266
  12. 2019What Do Single-view 3D Reconstruction Networks Learn?cited by 259
38 more
  1. 2019Learning Shape Templates with Structured Implicit Functionscited by 242
  2. 2019BSP-Net: Generating Compact Meshes via Binary Space Partitioningcited by 205
  3. 2016Thingi10K: A Dataset of 10,000 3D-Printing Modelscited by 182
  4. 2016Learning Shape Abstractions by Assembling Volumetric Primitivescited by 181
  5. 2019CvxNet: Learnable Convex Decompositioncited by 173
  6. 2019Superquadrics Revisited: Learning 3D Shape Parsing beyond Cuboidscited by 123
  7. 2017CSGNet: Neural Shape Parser for Constructive Solid Geometrycited by 106
  8. 2023Neural Kernel Surface Reconstructioncited by 102
  9. 2021Neural Parts: Learning Expressive 3D Shape Abstractions with Invertible Neural Networkscited by 101
  10. 2020ParSeNet: A Parametric Surface Fitting Network for 3D Point Cloudscited by 96
  11. 20223DILG: Irregular Latent Grids for 3D Generative Modelingcited by 92
  12. 2022POCO: Point Convolution for Surface Reconstructioncited by 91
  13. 2018Supervised Fitting of Geometric Primitives to 3D Point Cloudscited by 90
  14. 2020DualSDF: Semantic Shape Manipulation using a Two-Level Representationcited by 82
  15. 2020PatchNets: Patch-Based Generalizable Deep Implicit 3D Shape Representationscited by 81
  16. 20173D-PRNN: Generating Shape Primitives with Recurrent Neural Networkscited by 77
  17. 2020UCSG-Net -- Unsupervised Discovering of Constructive Solid Geometry Treecited by 77
  18. 2020Learning Unsupervised Hierarchical Part Decomposition of 3D Objects from a Single RGB Imagecited by 72
  19. 2021CAPRI-Net: Learning Compact CAD Shapes with Adaptive Primitive Assemblycited by 62
  20. 2021SA-ConvONet: Sign-Agnostic Optimization of Convolutional Occupancy Networkscited by 59
  21. 2021Point2Cyl: Reverse Engineering 3D Objects from Point Clouds to Extrusion Cylinderscited by 59
  22. 2021Neural Fields as Learnable Kernels for 3D Reconstructioncited by 58
  23. 2021CSG-Stump: A Learning Friendly CSG-Like Representation for Interpretable Shape Parsingcited by 51
  24. 2023SECAD-Net: Self-Supervised CAD Reconstruction by Learning Sketch-Extrude Operationscited by 46
  25. 2021Unsupervised Learning for Cuboid Shape Abstraction via Joint Segmentation from Point Cloudscited by 44
  26. 2023Animal3D: A Comprehensive Dataset of 3D Animal Pose and Shapecited by 43
  27. 2020D2IM-Net: Learning Detail Disentangled Implicit Fields from Single Imagescited by 42
  28. 2023D2CSG: Unsupervised Learning of Compact CSG Trees with Dual Complements and Dropoutscited by 39
  29. 2022ExtrudeNet: Unsupervised Inverse Sketch-and-Extrude for Shape Parsingcited by 38
  30. 2019Deep Parametric Shape Predictions using Distance Fieldscited by 30
  31. 2021Robust and Accurate Superquadric Recovery: a Probabilistic Approachcited by 29
  32. 2023Marching-Primitives: Shape Abstraction from Signed Distance Functioncited by 18
  33. 2020Neural Star Domain as Primitive Representationcited by 17
  34. 2022Primitive-based Shape Abstraction via Nonparametric Bayesian Inferencecited by 14
  35. 2021AIR-Nets: An Attention-Based Framework for Locally Conditioned Implicit Representationscited by 13
  36. 2022OReX: Object Reconstruction from Planar Cross-sections Using Neural Fieldscited by 7
  37. 2021PriFit: Learning to Fit Primitives Improves Few Shot Point Cloud Segmentationcited by 6
  38. 2023Neural Parametric Surfaces for Shape Modelingcited by 5

50 most cited of 51 All 51 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 2; 2026: 3.

5 papers.

  1. 2025Residual Primitive Fitting of 3D Shapes with SuperFrustacited by 5
  2. 2025Advances in 4D Representation: Geometry, Motion, and Interactioncited by 4
  3. 2026Learning Convex Decomposition via Feature Fieldscited by 1
  4. 2026Prompting Image Generators for Training-free Primitive Shape Abstractioncited by 0
  5. 2026Task-Adaptive Grounded 3D-Programmers Using 2D VLMscited by 0