mlsanityData to 14 Oct 2026

Attend, Infer, Repeat: Fast Scene Understanding with Generative Models

arXiv 1603.08575 ↗First posted 28 Mar 2016cs.CV cs.LGcited by 209

Not on the map. Why?

Abstract (arXiv)

We present a framework for efficient inference in structured image models that explicitly reason about objects. We achieve this by performing probabilistic inference using a recurrent neural network that attends to scene elements and processes them one at a time. Crucially, the model itself learns to choose the appropriate number of inference steps. We use this scheme to learn to perform inference in partially specified 2D models (variable-sized variational auto-encoders) and fully specified 3D models (probabilistic renderers). We show that such models learn to identify multiple objects - counting, locating and classifying the elements of a scene - without any supervision, e.g., decomposing 3D images with various numbers of objects in a single forward pass of a neural network. We further show that the networks produce accurate inferences when compared to supervised counterparts, and that their structure leads to improved generalization.

Key facts. Cited by 209 papers in the map (data through 14 Oct 2026). First posted on arXiv 28 Mar 2016.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 67; 2022: 47; 2023: 42; 2024: 26; 2025: 12; 2026: 15.

209 papers.

Sort:

By area:

  1. 2021Dynamic Neural Networks: A Surveycited by 408
  2. 2021Conditional Object-Centric Learning from Videocited by 248
  3. 2022Bridging the Gap to Real-World Object-Centric Learningcited by 220
  4. 2021Illiterate DALL-E Learns to Composecited by 147
  5. 2022Simple Unsupervised Object-Centric Learning for Complex and Naturalistic Videoscited by 139
  6. 2021Generative Adversarial Transformerscited by 120
  7. 2021GENESIS-V2: Inferring Unordered Object Representations without Iterative Refinementcited by 109
  8. 2021Unsupervised Discovery of Object Radiance Fieldscited by 101
  9. 2022Deep Image Deblurring: A Surveycited by 94
  10. 2021ClevrTex: A Texture-Rich Benchmark for Unsupervised Multi-Object Segmentationcited by 93
  11. 2023Object-Centric Slot Diffusioncited by 90
  12. 2021Decomposing 3D Scenes into Objects via Unsupervised Volume Segmentationcited by 88
38 more
  1. 2023SlotDiffusion: Object-Centric Generative Modeling with Diffusion Modelscited by 85
  2. 2021Generalization and Robustness Implications in Object-Centric Learningcited by 81
  3. 2021SetVAE: Learning Hierarchical Composition for Generative Modeling of Set-Structured Datacited by 78
  4. 2021Is Attention Better Than Matrix Decomposition?cited by 77
  5. 2021Neural Production Systems: Learning Rule-Governed Visual Dynamicscited by 75
  6. 2021Unsupervised Learning of Compositional Energy Conceptscited by 71
  7. 2021SIMONe: View-Invariant, Temporally-Abstracted Object Representations via Unsupervised Video Decompositioncited by 69
  8. 2021Differentiable Patch Selection for Image Recognitioncited by 68
  9. 2022Improving Object-centric Learning with Query Optimizationcited by 60
  10. 2021SemanticStyleGAN: Learning Compositional Generative Priors for Controllable Image Synthesis and Editingcited by 58
  11. 2023Additive Decoders for Latent Variables Identification and Cartesian-Product Extrapolationcited by 58
  12. 2023Invariant Slot Attention: Object Discovery with Slot-Centric Reference Framescited by 51
  13. 2021Unsupervised Layered Image Decomposition into Object Prototypescited by 50
  14. 2023An Investigation into Pre-Training Object-Centric Representations for Reinforcement Learningcited by 49
  15. 2023Self-supervised Object-Centric Learning for Videoscited by 49
  16. 2023SPOT: Self-Training with Patch-Order Permutation for Object-Centric Learning with Autoregressive Transformerscited by 49
  17. 2021Learning Disentangled Representations in the Imaging Domaincited by 46
  18. 2022Discovering Objects that Can Movecited by 44
  19. 2022Complex-Valued Autoencoders for Object Discoverycited by 40
  20. 2022Deep Learning Methods for Abstract Visual Reasoning: A Survey on Raven's Progressive Matricescited by 37
  21. 2022Neural Systematic Bindercited by 37
  22. 2021Efficient Iterative Amortized Inference for Learning Symmetric and Disentangled Multi-Object Representationscited by 36
  23. 2021MarioNette: Self-Supervised Sprite Learningcited by 35
  24. 2021From Machine Learning to Robotics: Challenges and Opportunities for Embodied Intelligencecited by 35
  25. 2022What to look at and where: Semantic and Spatial Refined Transformer for detecting human-object interactionscited by 35
  26. 2024Adaptive Slot Attention: Object Discovery with Dynamic Slot Numbercited by 34
  27. 2022VAEL: Bridging Variational Autoencoders and Probabilistic Logic Programmingcited by 31
  28. 2023OCAtari: Object-Centric Atari 2600 Reinforcement Learning Environmentscited by 30
  29. 2021Unsupervised Object Learning via Common Fatecited by 29
  30. 2023Rotating Features for Object Discoverycited by 29
  31. 2022Unsupervised Discovery and Composition of Object Light Fieldscited by 28
  32. 2022Compositional Scene Representation Learning via Reconstruction: A Surveycited by 27
  33. 2021Unsupervised Foreground Extraction via Deep Region Competitioncited by 26
  34. 20213DP3: 3D Scene Perception via Probabilistic Programmingcited by 26
  35. 2022OGC: Unsupervised 3D Object Segmentation from Rigid Dynamics of Point Cloudscited by 25
  36. 2021Variable-rate discrete representation learningcited by 24
  37. 2021Generative Video Transformer: Can Objects be the Words?cited by 24
  38. 2021Compositional Transformers for Scene Generationcited by 24

50 most cited of 209 All 209 (JSON)