mlsanityData to 14 Oct 2026

Unsupervised Learning of 3D Structure from Images

arXiv 1607.00662 ↗First posted 3 Jul 2016cs.CV cs.LG stat.MLcited by 67

Not on the map. Why?

Abstract (arXiv)

A key goal of computer vision is to recover the underlying 3D structure from 2D observations of the world. In this paper we learn strong deep generative models of 3D structures, and recover these structures from 3D and 2D images via probabilistic inference. We demonstrate high-quality samples and report log-likelihoods on several datasets, including ShapeNet [2], and establish the first benchmarks in the literature. We also show how these models and their inference networks can be trained end-to-end from 2D images. This demonstrates for the first time the feasibility of learning to infer 3D representations of the world in a purely unsupervised manner.

Key facts. Cited by 67 papers in the map (data through 14 Oct 2026). First posted on arXiv 3 Jul 2016.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 23; 2022: 18; 2023: 20; 2024: 3; 2025: 1; 2026: 2.

67 papers.

Sort:

By area:

  1. 2021Common Objects in 3D: Large-Scale Learning and Evaluation of Real-life 3D Category Reconstructioncited by 652
  2. 2021Light Field Networks: Neural Scene Representations with Single-Evaluation Renderingcited by 332
  3. 2021StyleSDF: High-Resolution 3D-Consistent Image and Geometry Generationcited by 289
  4. 2023MeshGPT: Generating Triangle Meshes with Decoder-Only Transformerscited by 259
  5. 20213D Neural Scene Representations for Visuomotor Controlcited by 111
  6. 2021Neural Parts: Learning Expressive 3D Shape Abstractions with Invertible Neural Networkscited by 103
  7. 2021gradSim: Differentiable simulation for system identification and visuomotor controlcited by 100
  8. 20213D Semantic Scene Completion: a Surveycited by 74
  9. 2021Unsupervised Learning of 3D Object Categories from Videos in the Wildcited by 67
  10. 2023Deep Learning on Implicit Neural Representations of Shapescited by 63
  11. 2022ShapeCrafter: A Recursive Text-Conditioned 3D Shape Generation Modelcited by 62
  12. 2021CAMPARI: Camera-Aware Decomposed Generative Neural Radiance Fieldscited by 55
38 more
  1. 2022GIFS: Neural Implicit Function for General Shape Representationcited by 45
  2. 2023Neural Volumetric Memory for Visual Locomotion Controlcited by 44
  3. 2024Where Do We Stand with Implicit Neural Representations? A Technical and Performance Surveycited by 42
  4. 2021Learning Object-Centric Representations of Multi-Object Scenes from Multiple Viewscited by 37
  5. 2021Video Autoencoder: self-supervised disentanglement of static 3D structure and motioncited by 34
  6. 20213D-Aware Semantic-Guided Generative Model for Human Synthesiscited by 31
  7. 2022DoF-NeRF: Depth-of-Field Meets Neural Radiance Fieldscited by 26
  8. 2023GINA-3D: Learning to Generate Implicit Neural Assets in the Wildcited by 25
  9. 2023AI-Generated Content (AIGC) for Various Data Modalities: A Surveycited by 25
  10. 2022ALTO: Alternating Latent Topologies for Implicit 3D Reconstructioncited by 21
  11. 2022SPARF: Large-Scale Learning of 3D Sparse Radiance Fields from Few Input Imagescited by 17
  12. 2023Decoupling Dynamic Monocular Videos for Dynamic View Synthesiscited by 17
  13. 2021NeeDrop: Self-supervised Shape Representation from Sparse Point Clouds using Needle Droppingcited by 16
  14. 2023Neural Vector Fields: Implicit Representation by Explicit Learningcited by 16
  15. 2022Primitive-based Shape Abstraction via Nonparametric Bayesian Inferencecited by 15
  16. 2024PASTA: Controllable Part-Aware Shape Generation with Autoregressive Transformerscited by 15
  17. 2021DeepMesh: Differentiable Iso-Surface Extractioncited by 14
  18. 2023Sketch-A-Shape: Zero-Shot Sketch-to-3D Shape Generationcited by 14
  19. 2022SymmNeRF: Learning to Explore Symmetry Prior for Single-View View Synthesiscited by 13
  20. 2023DeFormer: Integrating Transformers with Deformable Models for 3D Shape Abstraction from a Single Imagecited by 12
  21. 2022MonoNeRF: Learning Generalizable NeRFs from Monocular Videos without Camera Posecited by 10
  22. 2022Multi-View Reconstruction using Signed Ray Distance Functions (SRDF)cited by 9
  23. 2021Unsupervised Shape Completion via Deep Prior in the Neural Tangent Kernel Perspectivecited by 8
  24. 2022Neural Color Operators for Sequential Image Retouchingcited by 8
  25. 2022Controllable Radiance Fields for Dynamic Face Synthesiscited by 8
  26. 2021An Effective Loss Function for Generating 3D Models from Single 2D Image without Renderingcited by 7
  27. 2021LightSAL: Lightweight Sign Agnostic Learning for Implicit Surface Representationcited by 7
  28. 2021Hierarchical View Predictor: Unsupervised 3D Global Feature Learning through Hierarchical Prediction among Unordered Viewscited by 7
  29. 2022NeuralODF: Learning Omnidirectional Distance Fields for 3D Shape Representationcited by 7
  30. 2022MVTN: Learning Multi-View Transformations for 3D Understandingcited by 7
  31. 2022Learning A Locally Unified 3D Point Cloud for View Synthesiscited by 6
  32. 2023Block-local learning with probabilistic latent representationscited by 6
  33. 2024Dynamic Neural Radiance Field From Defocused Monocular Videocited by 6
  34. 2021Single View Physical Distance Estimation using Human Posecited by 5
  35. 2026MeshFlow: Mesh Generation with Equivariant Flow Matchingcited by 4
  36. 2022Exploiting Inductive Bias in Transformers for Unsupervised Disentanglement of Syntax and Semantics with VAEscited by 3
  37. 2022Near out-of-distribution detection for low-resolution radar micro-Doppler signaturescited by 3
  38. 2023Inferring Fluid Dynamics via Inverse Renderingcited by 3

50 most cited of 67 All 67 (JSON)