mlsanityData to 7 Oct 2026

Superquadrics Revisited: Learning 3D Shape Parsing beyond Cuboids

arXiv 1904.09970 ↗First posted 22 Apr 2019cs.CVcited by 123

Not on the map. Why?

Abstract (arXiv)

Abstracting complex 3D shapes with parsimonious part-based representations has been a long standing goal in computer vision. This paper presents a learning-based solution to this problem which goes beyond the traditional 3D cuboid representation by exploiting superquadrics as atomic elements. We demonstrate that superquadrics lead to more expressive 3D scene parses while being easier to learn than 3D cuboid representations. Moreover, we provide an analytical solution to the Chamfer loss which avoids the need for computational expensive reinforcement learning or iterative prediction. Our model learns to parse 3D objects into consistent superquadric representations without supervision. Results on various ShapeNet categories as well as the SURREAL human body dataset demonstrate the flexibility of our model in capturing fine details and complex poses that could not have been modelled using cuboids.

Key facts. Cited by 123 papers in the map (data through 7 Oct 2026). First posted on arXiv 22 Apr 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 30; 2022: 23; 2023: 23; 2024: 15; 2025: 15; 2026: 17.

123 papers.

Sort:

By area:

  1. 2023LucidDreamer: Domain-free Generation of 3D Gaussian Splatting Scenescited by 228
  2. 2021ATISS: Autoregressive Transformers for Indoor Scene Synthesiscited by 195
  3. 2021Im2Vec: Synthesizing Vector Graphics without Vector Supervisioncited by 114
  4. 2021Neural Parts: Learning Expressive 3D Shape Abstractions with Invertible Neural Networkscited by 101
  5. 2021Holistic 3D Scene Understanding from a Single Image with Implicit Representationcited by 94
  6. 2023SALAD: Part-Level Latent Diffusion for 3D Shape Generation and Manipulationcited by 66
  7. 2021CAPRI-Net: Learning Compact CAD Shapes with Adaptive Primitive Assemblycited by 62
  8. 2022Approximate Convex Decomposition for 3D Meshes with Collision-Aware Concavity and Tree Searchcited by 61
  9. 2021Point2Cyl: Reverse Engineering 3D Objects from Point Clouds to Extrusion Cylinderscited by 59
  10. 2024Part123: Part-aware 3D Reconstruction from a Single-view Imagecited by 54
  11. 2021CSG-Stump: A Learning Friendly CSG-Like Representation for Interpretable Shape Parsingcited by 51
  12. 20243D Convex Splatting: Radiance Field Rendering with 3D Smooth Convexescited by 50
38 more
  1. 2023SECAD-Net: Self-Supervised CAD Reconstruction by Learning Sketch-Extrude Operationscited by 46
  2. 2021Unsupervised Learning for Cuboid Shape Abstraction via Joint Segmentation from Point Cloudscited by 44
  3. 2022Towards High-Fidelity Single-view Holistic Reconstruction of Indoor Scenescited by 43
  4. 2023D2CSG: Unsupervised Learning of Compact CSG Trees with Dual Complements and Dropoutscited by 39
  5. 2022ExtrudeNet: Unsupervised Inverse Sketch-and-Extrude for Shape Parsingcited by 38
  6. 2023Differentiable Blocks World: Qualitative 3D Decomposition by Rendering Primitivescited by 37
  7. 2023PartNeRF: Generating Part-Aware Editable 3D Shapes without 3D Supervisioncited by 36
  8. 2022PartAfford: Part-level Affordance Discovery from 3D Objectscited by 35
  9. 2022Neural Template: Topology-aware Reconstruction and Disentangled Generation of 3D Meshescited by 34
  10. 2024Img2CAD: Reverse Engineering 3D CAD Models from Images through VLM-Assisted Conditional Factorizationcited by 34
  11. 2021PartGlot: Learning Shape Part Segmentation from Language Reference Gamescited by 30
  12. 2021Robust and Accurate Superquadric Recovery: a Probabilistic Approachcited by 29
  13. 2021CPFN: Cascaded Primitive Fitting Networks for High-Resolution Point Cloudscited by 24
  14. 2021ODAM: Object Detection, Association, and Mapping using Posed RGB Videocited by 22
  15. 2021Learning 3D Dense Correspondence via Canonical Point Autoencodercited by 21
  16. 2021EditVAE: Unsupervised Part-Aware Controllable 3D Point Cloud Shape Generationcited by 21
  17. 2022Pixel2Mesh++: 3D Mesh Generation and Refinement from Multi-View Imagescited by 21
  18. 20243Doodle: Compact Abstraction of Objects with 3D Strokescited by 21
  19. 2025SuperDec: 3D Scene Decomposition with Superquadric Primitivescited by 19
  20. 2022RIM-Net: Recursive Implicit Fields for Unsupervised Learning of Hierarchical Shape Structurescited by 18
  21. 2023Marching-Primitives: Shape Abstraction from Signed Distance Functioncited by 18
  22. 2025C3G: Learning Compact 3D Representations with 2K Gaussianscited by 18
  23. 2021Cuboids Revisited: Learning Robust 3D Shape Fitting to Single RGB Imagescited by 17
  24. 20213DIAS: 3D Shape Reconstruction with Implicit Algebraic Surfacescited by 17
  25. 20213D Compositional Zero-shot Learning with DeCompositional Consensuscited by 16
  26. 20243D Part Segmentation via Geometric Aggregation of 2D Visual Featurescited by 15
  27. 2022Primitive-based Shape Abstraction via Nonparametric Bayesian Inferencecited by 14
  28. 2023Doodle Your 3D: From Abstract Freehand Sketches to Precise 3D Shapescited by 14
  29. 2024PASTA: Controllable Part-Aware Shape Generation with Autoregressive Transformerscited by 14
  30. 2025CasaGPT: Cuboid Arrangement and Scene Assembly for Interior Designcited by 14
  31. 2025GLOVER++: Unleashing the Potential of Affordance Learning from Human Behaviors for Robotic Manipulationcited by 14
  32. 2021Discovering 3D Parts from Image Collectionscited by 13
  33. 2023Iterative Superquadric Recomposition of 3D Objects from Multiple Viewscited by 13
  34. 2021Unsupervised Pose-Aware Part Decomposition for 3D Articulated Objectscited by 12
  35. 2021DeepCurrents: Learning Implicit Representations of Shapes with Boundariescited by 12
  36. 2025PrimitiveAnything: Human-Crafted 3D Primitive Assembly Generation with Auto-Regressive Transformercited by 12
  37. 2022MonteBoxFinder: Detecting and Filtering Primitives to Fit a Noisy Point Cloudcited by 11
  38. 2023DeFormer: Integrating Transformers with Deformable Models for 3D Shape Abstraction from a Single Imagecited by 11

50 most cited of 123 All 123 (JSON)