mlsanityData to 7 Oct 2026

Unsupervised Discovery of 3D Physical Objects from Video

arXiv 2007.12348 ↗First posted 24 Jul 2020cs.CV cs.LGcited by 27

Not on the map. Why?

Abstract (arXiv)

We study the problem of unsupervised physical object discovery. While existing frameworks aim to decompose scenes into 2D segments based off each object's appearance, we explore how physics, especially object interactions, facilitates disentangling of 3D geometry and position of objects from video, in an unsupervised manner. Drawing inspiration from developmental psychology, our Physical Object Discovery Network (POD-Net) uses both multi-scale pixel cues and physical motion cues to accurately segment observable and partially occluded objects of varying sizes, and infer properties of those objects. Our model reliably segments objects on both synthetic and real scenes. The discovered object properties can also be used to reason about physical events.

Key facts. Cited by 27 papers in the map (data through 7 Oct 2026). First posted on arXiv 24 Jul 2020.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2021: 3; 2022: 9; 2023: 8; 2024: 6; 2025: 1.

27 papers.

Sort:

By area:

  1. 2021Conditional Object-Centric Learning from Videocited by 247
  2. 2022Simple Unsupervised Object-Centric Learning for Complex and Naturalistic Videoscited by 139
  3. 2023Object-Centric Slot Diffusioncited by 90
  4. 2021Neural Production Systems: Learning Rule-Governed Visual Dynamicscited by 75
  5. 2021Unsupervised Learning of Compositional Energy Conceptscited by 71
  6. 2022Discovering Objects that Can Movecited by 44
  7. 2022Learning to Detect Mobile Objects from LiDAR Scans Without Labelscited by 44
  8. 2022Neural Systematic Bindercited by 37
  9. 2023Unsupervised Compositional Concepts Discovery with Text-to-Image Generative Modelscited by 36
  10. 2023Rotating Features for Object Discoverycited by 29
  11. 2022Compositional Scene Representation Learning via Reconstruction: A Surveycited by 27
  12. 2023Object Discovery from Motion-Guided Tokenscited by 25
15 more
  1. 2022Neural Groundplans: Persistent Neural Scene Representations from a Single Imagecited by 17
  2. 2022Boosting Object Representation Learning via Motion and Object Continuitycited by 17
  3. 2023Learning 3D Particle-based Simulators from RGB-D Videoscited by 17
  4. 2024Learning Discrete Concepts in Latent Hierarchical Modelscited by 17
  5. 2022Unsupervised Segmentation in Real-World Images via Spelke Object Inferencecited by 15
  6. 2024Compositional Image Decomposition with Diffusion Modelscited by 12
  7. 2022Test-time Adaptation with Slot-Centric Modelscited by 9
  8. 2024Neural Language of Thought Modelscited by 9
  9. 2023Benchmarks for Physical Reasoning AIcited by 8
  10. 2023RobustCLEVR: A Benchmark and Framework for Evaluating Robustness in Object-centric Learningcited by 6
  11. 2023Perceiving Unseen 3D Objects by Poking the Objectscited by 5
  12. 2024Binding Dynamics in Rotating Featurescited by 5
  13. 2025Nonparametric Identification of Latent Conceptscited by 3
  14. 2024Parallelized Spatiotemporal Bindingcited by 2
  15. 2024MOD-UV: Learning Mobile Object Detectors from Unlabeled Videoscited by 2