mlsanityData to 7 Oct 2026

Neural Eulerian Scene Flow Fields

arXiv 2410.02031 ↗First posted 2 Oct 2024cs.CVcited by 14

Not on the map. Why?

Abstract (arXiv)

We reframe scene flow as the task of estimating a continuous space-time ODE that describes motion for an entire observation sequence, represented with a neural prior. Our method, EulerFlow, optimizes this neural prior estimate against several multi-observation reconstruction objectives, enabling high quality scene flow estimation via pure self-supervision on real-world data. EulerFlow works out-of-the-box without tuning across multiple domains, including large-scale autonomous driving scenes and dynamic tabletop settings. Remarkably, EulerFlow produces high quality flow estimates on small, fast moving objects like birds and tennis balls, and exhibits emergent 3D point tracking behavior by solving its estimated ODE over long-time horizons. On the Argoverse 2 2024 Scene Flow Challenge, EulerFlow outperforms all prior art, surpassing the next-best unsupervised method by more than 2.5x, and even exceeding the next-best supervised method by over 10%.

Key facts. Cited by 14 papers in the map (data through 7 Oct 2026). Builds on 40 of its 51 references that are in the map. First posted on arXiv 2 Oct 2024.

Builds on

40 of 51 references.

References · 40

  1. 2018Neural Ordinary Differential Equationscited by 3,733
  2. 2019Scalability in Perception for Autonomous Driving: Waymo Open Datasetcited by 2,436
  3. 2020Nerfies: Deformable Neural Radiance Fieldscited by 1,194
  4. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 878
  5. 2020Accelerating 3D Deep Learning with PyTorch3Dcited by 781
  6. 2023Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecastingcited by 701
  7. 2019Class-balanced Grouping and Sampling for Point Cloud 3D Object Detectioncited by 350
  8. 2018FlowNet3D: Learning Scene Flow in 3D Point Cloudscited by 241
  9. 2024DepthCrafter: Generating Consistent Long Depth Sequences for Open-world Videoscited by 202
  10. 2020Liquid Time-constant Networkscited by 160
  11. 2020FLOT: Scene Flow on Point Clouds Guided by Optimal Transportcited by 129
  12. 2019HPLFlowNet: Hierarchical Permutohedral Lattice FlowNet for Scene Flow Estimation on Large-scale Point Cloudscited by 108
28 more
  1. 2021Neural Scene Flow Priorcited by 91
  2. 2022GARF: Gaussian Activated Radiance Fields for High Fidelity Reconstruction and Pose Estimationcited by 86
  3. 2023Point Cloud Forecasting as a Proxy for 4D Occupancy Forecastingcited by 76
  4. 2020FlowStep3D: Model Unrolling for Self-Supervised Scene Flow Estimationcited by 71
  5. 2022Motion Inspired Unsupervised Perception and Prediction in Autonomous Drivingcited by 59
  6. 2021FabricFlowNet: Bimanual Cloth Manipulation with a Flow-based Policycited by 54
  7. 2021Scalable Scene Flow from Point Clouds in the Real Worldcited by 49
  8. 2020Self-Supervised Learning of Non-Rigid Residual Flow and Ego-Motioncited by 41
  9. 2022Towards Long-Tailed 3D Detectioncited by 40
  10. 2024UnO: Unsupervised Occupancy Fields for Perception and Forecastingcited by 40
  11. 2023Fast Neural Scene Flowcited by 39
  12. 2024SeFlow: A Self-Supervised Scene Flow Method in Autonomous Drivingcited by 38
  13. 2021HCRF-Flow: Scene Flow from Point Clouds with Continuous High-order CRFs and Position-aware Flow Embeddingcited by 36
  14. 2024ICP-Flow: LiDAR Scene Flow Estimation with ICPcited by 34
  15. 2024DeFlow: Decoder of Scene Flow Network in Autonomous Drivingcited by 29
  16. 2022Forecasting from LiDAR via Future Object Detectioncited by 28
  17. 2024Flow4D: Leveraging 4D Voxel Network for LiDAR Scene Flow Estimationcited by 22
  18. 2024I Can't Believe It's Not Scene Flow!cited by 21
  19. 20223D Object Detection with a Self-supervised Lidar Scene Flow Backbonecited by 19
  20. 2022RMS-FlowNet: Efficient and Robust Multi-Scale Scene Flow Estimation for Large-Scale Point Cloudscited by 13
  21. 2023ZeroFlow: Scalable Scene Flow via Distillationcited by 13
  22. 2023Multi-Body Neural Scene Flowcited by 12
  23. 2021Sparse PointPillars: Maintaining and Exploiting Input Sparsity to Improve Runtime on Embedded Systemscited by 11
  24. 2020FlowMOT: 3D Multi-Object Tracking by Scene Flow Associationcited by 10
  25. 2021GEASI: Geodesic-based Earliest Activation Sites Identification in cardiac modelscited by 9
  26. 2023An Empirical Analysis of Range for 3D Object Detectioncited by 9
  27. 2022A Brief Survey on Person Recognition at a Distancecited by 4
  28. 2024Self-Supervised Multi-Frame Neural Scene Flowcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 6; 2026: 8.

14 papers.

Sort:

By area:

  1. 2025Any4D: Unified Feed-Forward Metric 4D Reconstructioncited by 32
  2. 2025HiMo: High-Speed Objects Motion Compensation in Point Cloudscited by 11
  3. 2025DeltaFlow: An Efficient Multi-frame Scene Flow Estimation Methodcited by 7
  4. 2025Flux4D: Flow-based Unsupervised 4D Reconstructioncited by 7
  5. 2025E-MoFlow: Learning Egomotion and Optical Flow from Event Data via Implicit Regularizationcited by 2
  6. 2026SynFlow: Scaling Up LiDAR Scene Flow Estimation with Synthetic Datacited by 1
  7. 2025UniFlow: Zero-Shot LiDAR Scene Flow for Autonomous Vehiclescited by 0
  8. 2026Weakly Supervised Cross-Modal Learning for 4D Radar Scene Flow Estimationcited by 0
  9. 2026H-Flow: Self-supervised Human Scene Flow via Physics-inspired Joint Multi-modal Learningcited by 0
  10. 2026ODeform: Learning Continuous 4D Motion for Shape Deformation with Neural ODEscited by 0
  11. 2026CorrelationFlow: A Training-Free Geometric Approach for LiDAR Scene Flow Estimationcited by 0
  12. 2026RVLoss: Runoff Vote Loss for Self-Supervised LiDAR Scene Flow Estimationcited by 0
2 more
  1. 2026TriO: Tri-Modal Unsupervised Occupancy World Model for Anything Perceptioncited by 0
  2. 2026A Differentiable Optimization Framework for Registering Sequential Bounding Boxes with Point Cloud Streamcited by 0