mlsanityData to 7 Oct 2026

Inductive Biases for Object-Centric Representations in the Presence of Complex Textures

arXiv 2204.08479 ↗First posted 18 Apr 2022cs.CV cs.LGcited by 13

Not on the map. Why?

Abstract (arXiv)

Understanding which inductive biases could be helpful for the unsupervised learning of object-centric representations of natural scenes is challenging. In this paper, we systematically investigate the performance of two models on datasets where neural style transfer was used to obtain objects with complex textures while still retaining ground-truth annotations. We find that by using a single module to reconstruct both the shape and visual appearance of each object, the model learns more useful representations and achieves better object separation. In addition, we observe that adjusting the latent space size is insufficient to improve segmentation performance. Finally, the downstream usefulness of the representations is significantly more strongly correlated with segmentation quality than with reconstruction accuracy.

Key facts. Cited by 13 papers in the map (data through 7 Oct 2026). Builds on 47 of its 71 references that are in the map. First posted on arXiv 18 Apr 2022.

Builds on

47 of 71 references.

References · 47

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  2. 2013Auto-Encoding Variational Bayescited by 3,980
  3. 2016CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoningcited by 1,445
  4. 2018Relational inductive biases, deep learning, and graph networkscited by 1,341
  5. 2019Dota 2 with Large Scale Deep Reinforcement Learningcited by 1,084
  6. 2018Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representationscited by 929
  7. 2020Learning to Simulate Complex Physics with Graph Networkscited by 867
  8. 2020Object-Centric Learning with Slot Attentioncited by 865
  9. 2016Interaction Networks for Learning about Objects, Relations and Physicscited by 503
  10. 2015A Neural Algorithm of Artistic Stylecited by 480
  11. 2019MONet: Unsupervised Scene Decomposition and Representationcited by 388
  12. 2019The Neuro-Symbolic Concept Learner: Interpreting Scenes, Words, and Sentences From Natural Supervisioncited by 369
35 more
  1. 2019Multi-Object Representation Learning with Iterative Variational Inferencecited by 359
  2. 2020ReZero is All You Need: Fast Convergence at Large Depthcited by 260
  3. 2021Conditional Object-Centric Learning from Videocited by 247
  4. 2020On the Binding Problem in Artificial Neural Networkscited by 230
  5. 2019GENESIS: Generative Scene Inference and Sampling with Object-Centric Latent Representationscited by 227
  6. 2019Recurrent Independent Mechanismscited by 219
  7. 2017Learning Explanatory Rules from Noisy Datacited by 210
  8. 2016Attend, Infer, Repeat: Fast Scene Understanding with Generative Modelscited by 209
  9. 2019Contrastive Learning of Structured World Modelscited by 206
  10. 2020SPACE: Unsupervised Object-Oriented Scene Representation via Spatial Attention and Decompositioncited by 190
  11. 2018Sequential Attend, Infer, Repeat: Generative Modelling of Moving Objectscited by 128
  12. 2017Neural Expectation Maximizationcited by 123
  13. 2021GENESIS-V2: Inferring Unordered Object Representations without Iterative Refinementcited by 109
  14. 2019SCALOR: Generative World Models with Scalable Object Representationscited by 108
  15. 2018Relational Neural Expectation Maximization: Unsupervised Discovery of Objects and their Interactionscited by 100
  16. 2019Are Disentangled Representations Helpful for Abstract Visual Reasoning?cited by 100
  17. 2020On Disentangled Representations Learned From Correlated Datacited by 94
  18. 2021ClevrTex: A Texture-Rich Benchmark for Unsupervised Multi-Object Segmentationcited by 93
  19. 2021Generalization and Robustness Implications in Object-Centric Learningcited by 81
  20. 2019Unsupervised Object Segmentation by Redrawingcited by 74
  21. 2020Improving Generative Imagination in Object-Centric World Modelscited by 70
  22. 2016Tagger: Deep Unsupervised Perceptual Groupingcited by 69
  23. 2020On the Transfer of Disentangled Representations in Realistic Settingscited by 63
  24. 2017Classical Planning in Deep Latent Space: Bridging the Subsymbolic-Symbolic Boundarycited by 58
  25. 2019Structured Object-Aware Physics Prediction for Video Modeling and Planningcited by 42
  26. 2019Exploiting Spatial Invariance for Scalable Unsupervised Object Trackingcited by 40
  27. 2020Towards causal generative scene models via competition of expertscited by 37
  28. 2020Learning to Manipulate Individual Objects in an Imagecited by 22
  29. 2020Learning Object-Centric Video Models by Contrasting Setscited by 19
  30. 2022Compositional Multi-Object Reinforcement Learning with Linear Relation Networkscited by 19
  31. 2020Reconstruction Bottlenecks in Object-Centric Generative Modelscited by 17
  32. 2021The Role of Pretrained Representations for the OOD Generalization of Reinforcement Learning Agentscited by 12
  33. 2020Planning from Pixels in Atari with Learned Symbolic Representationscited by 11
  34. 2020Slot Contrastive Networks: A Contrastive Approach for Representing Objectscited by 9
  35. 2019LAVAE: Disentangling Location and Appearancecited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 5; 2024: 6; 2025: 0; 2026: 1.

13 papers.

Sort:

By area:

  1. 2023Provably Learning Object-Centric Representationscited by 45
  2. 2024Position: Understanding LLMs Requires More Than Statistical Generalizationcited by 20
  3. 2022Promising or Elusive? Unsupervised Object Segmentation from Real-world Single Imagescited by 16
  4. 2024Composing Pre-Trained Object-Centric Representations for Robotics From "What" and "Where" Foundation Modelscited by 16
  5. 2024Exploring the Effectiveness of Object-Centric Representations in Visual Question Answering: Comparative Insights with Foundation Modelscited by 14
  6. 2023Neural Radiance Field Codebookscited by 7
  7. 2024Recasting Generic Pretrained Vision Transformers As Object-Centric Scene Encoders For Manipulation Policiescited by 6
  8. 2023Benchmarking and Analysis of Unsupervised Object Segmentation from Real-world Single Imagescited by 4
  9. 2024Explicitly Disentangled Representations in Object-Centric Learningcited by 4
  10. 2023Exploring the Role of the Bottleneck in Slot-Based Models Through Covariance Regularizationcited by 3
  11. 2024OpenSlot: Mixed Open-Set Recognition with Object-Centric Learningcited by 2
  12. 2023Diffusion Based Causal Representation Learningcited by 0
1 more
  1. 2026Are Object-Centric Representations Better At Compositional Generalization?cited by 0