mlsanityData to 14 Oct 2026

Semantic Attention Flow Fields for Monocular Dynamic Scene Decomposition

arXiv 2303.01526 ↗First posted 2 Mar 2023cs.CVcited by 13

Not on the map. Why?

Abstract (arXiv)

From video, we reconstruct a neural volume that captures time-varying color, density, scene flow, semantics, and attention information. The semantics and attention let us identify salient foreground objects separately from the background across spacetime. To mitigate low resolution semantic and attention features, we compute pyramids that trade detail with whole-image context. After optimization, we perform a saliency-aware clustering to decompose the scene. To evaluate real-world scenes, we annotate object masks in the NVIDIA Dynamic Scene and DyCheck datasets. We demonstrate that this method can decompose dynamic scenes in an unsupervised way with competitive performance to a supervised method, and that it improves foreground/background segmentation over recent static/dynamic split methods. Project Webpage: https://visual.cs.brown.edu/saff

Key facts. Cited by 13 papers in the map (data through 14 Oct 2026). Builds on 50 of its 59 references that are in the map. First posted on arXiv 2 Mar 2023.

Builds on

50 of 59 references.

References · 50

  1. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  2. 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,723
  3. 2017Billion-scale similarity search with GPUscited by 2,497
  4. 2019Scalability in Perception for Autonomous Driving: Waymo Open Datasetcited by 2,451
  5. 2020RAFT: Recurrent All-Pairs Field Transforms for Optical Flowcited by 2,415
  6. 2021Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fieldscited by 1,964
  7. 2019Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfercited by 1,495
  8. 2022TensoRF: Tensorial Radiance Fieldscited by 1,280
  9. 2020Nerfies: Deformable Neural Radiance Fieldscited by 1,198
  10. 2020Object-Centric Learning with Slot Attentioncited by 876
  11. 2020GIRAFFE: Representing Scenes as Compositional Generative Neural Feature Fieldscited by 785
  12. 2020Neural Scene Flow Fields for Space-Time View Synthesis of Dynamic Scenescited by 751
38 more
  1. 2021BARF: Bundle-Adjusting Neural Radiance Fieldscited by 642
  2. 2021Neural Fields in Visual Computing and Beyondcited by 540
  3. 2020Non-Rigid Neural Radiance Fields: Reconstruction and Novel View Synthesis of a Dynamic Scene From Monocular Videocited by 471
  4. 2021In-Place Scene Labelling and Understanding with Implicit Scene Representationcited by 415
  5. 2019Video Instance Segmentationcited by 393
  6. 2019MONet: Unsupervised Scene Decomposition and Representationcited by 392
  7. 2022Kubric: A scalable dataset generatorcited by 381
  8. 2022Decomposing NeRF for Editing via Feature Field Distillationcited by 371
  9. 2019Multi-Object Representation Learning with Iterative Variational Inferencecited by 360
  10. 2021Deep ViT Features as Dense Visual Descriptorscited by 330
  11. 2021Conditional Object-Centric Learning from Videocited by 248
  12. 2022Panoptic Neural Fields: A Semantic Object-Aware Neural Scene Representationcited by 243
  13. 2021Learning Object-Compositional Neural Radiance Field for Editable Scene Renderingcited by 228
  14. 2022Bridging the Gap to Real-World Object-Centric Learningcited by 220
  15. 2022Monocular Dynamic View Synthesis: A Reality Checkcited by 200
  16. 2022Neural Feature Fusion Fields: 3D Distillation of Self-Supervised 2D Image Representationscited by 196
  17. 2020Robust Consistent Video Depth Estimationcited by 188
  18. 2022Self-Supervised Transformers for Unsupervised Object Discovery using Normalized Cutcited by 167
  19. 2020Novel View Synthesis of Dynamic Scenes with Globally Coherent Depths from a Monocular Cameracited by 166
  20. 2022SAVi++: Towards End-to-End Object-Centric Learning from Real-World Videoscited by 166
  21. 2022Panoptic NeRF: 3D-to-2D Label Transfer for Panoptic Urban Scene Segmentationcited by 152
  22. 2022Deep Spectral Methods: A Surprisingly Strong Baseline for Unsupervised Semantic Segmentation and Localizationcited by 150
  23. 2022D2NeRF: Self-Supervised Decoupling of Dynamic and Static Objects from a Monocular Videocited by 137
  24. 2019Entity Abstraction in Visual Model-Based Reinforcement Learningcited by 110
  25. 2021Unsupervised Discovery of Object Radiance Fieldscited by 101
  26. 2019The 2019 DAVIS Challenge on VOS: Unsupervised Multi-Object Segmentationcited by 97
  27. 2021Decomposing 3D Scenes into Objects via Unsupervised Volume Segmentationcited by 88
  28. 2021Video Instance Segmentation with a Propose-Reduce Paradigmcited by 74
  29. 2022NeRF-SOS: Any-View Self-supervised Object Segmentation on Complex Scenescited by 71
  30. 2021SIMONe: View-Invariant, Temporally-Abstracted Object Representations via Unsupervised Video Decompositioncited by 69
  31. 2022Deformable Sprites for Unsupervised Video Decompositioncited by 61
  32. 2021NeuralDiff: Segmenting 3D objects that move in egocentric videoscited by 57
  33. 2022Unsupervised Salient Object Detection with Spectral Cluster Votingcited by 56
  34. 2021Unsupervised Layered Image Decomposition into Object Prototypescited by 50
  35. 2022Feature-Realistic Neural Fusion for Real-Time, Open Set Scene Understandingcited by 37
  36. 2022Unsupervised Discovery and Composition of Object Light Fieldscited by 28
  37. 2022Revealing Occlusions with 4D Neural Fieldscited by 17
  38. 2022Baking in the Feature: Accelerating Volumetric Segmentation by Rendering Feature Mapscited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 2; 2024: 5; 2025: 5; 2026: 1.

13 papers.

Sort:

  1. 2023GauFRe: Gaussian Deformation Fields for Real-time Dynamic Novel View Synthesiscited by 95
  2. 2023EPIC Fields: Marrying 3D Geometry and Video Understandingcited by 49
  3. 2024Grid4D: 4D Decomposed Hash Encoding for High-Fidelity Dynamic Gaussian Splattingcited by 49
  4. 2024WorDepth: Variational Language Prior for Monocular Depth Estimationcited by 34
  5. 2024DGD: Dynamic 3D Gaussians Distillationcited by 23
  6. 2024When Does Perceptual Alignment Benefit Vision Representations?cited by 18
  7. 2024Semantically-aware Neural Radiance Fields for Visual Scene Understanding: A Comprehensive Reviewcited by 9
  8. 2025DASH: 4D Hash Encoding with Self-Supervised Decomposition for Real-Time Dynamic Scene Renderingcited by 8
  9. 2025Advances in Radiance Field for Dynamic Scene: From Neural Field to Gaussian Fieldcited by 5
  10. 2025DIV-FF: Dynamic Image-Video Feature Fields For Environment Understanding in Egocentric Videoscited by 0
  11. 2025Layered Motion Fusion: Lifting Motion Segmentation to 3D in Egocentric Videoscited by 0
  12. 2025Laplacian Analysis Meets Dynamics Modelling: Gaussian Splatting for 4D Reconstructioncited by 0
1 more
  1. 2026PUGS: Physics-Guided Uncertainty Modeling for Underwater Dynamic 3D Gaussian Splattingcited by 0