mlsanityData to 14 Oct 2026

FlowNet3D++: Geometric Losses For Deep Scene Flow Estimation

arXiv 1912.01438 ↗First posted 3 Dec 2019cs.CV cs.ROcited by 55

Not on the map. Why?

Abstract (arXiv)

We present FlowNet3D++, a deep scene flow estimation network. Inspired by classical methods, FlowNet3D++ incorporates geometric constraints in the form of point-to-plane distance and angular alignment between individual vectors in the flow field, into FlowNet3D. We demonstrate that the addition of these geometric loss terms improves the previous state-of-art FlowNet3D accuracy from 57.85% to 63.43%. To further demonstrate the effectiveness of our geometric constraints, we propose a benchmark for flow estimation on the task of dynamic 3D reconstruction, thus providing a more holistic and practical measure of performance than the breakdown of individual metrics previously used to evaluate scene flow. This is made possible through the contribution of a novel pipeline to integrate point-based scene flow predictions into a global dense volume. FlowNet3D++ achieves up to a 15.0% reduction in reconstruction error over FlowNet3D, and up to a 35.2% improvement over KillingFusion alone. We will release our scene flow estimation code later.

Key facts. Cited by 55 papers in the map (data through 14 Oct 2026). First posted on arXiv 3 Dec 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 14; 2022: 7; 2023: 13; 2024: 13; 2025: 5; 2026: 3.

55 papers.

Sort:

By area:

  1. 2024Shape of Motion: 4D Reconstruction from a Single Videocited by 250
  2. 2024SpatialTracker: Tracking Any 2D Pixels in 3D Spacecited by 181
  3. 20214DComplete: Non-Rigid Motion Estimation Beyond the Observable Surfacecited by 124
  4. 2021Neural Scene Flow Priorcited by 91
  5. 2025TAPIP3D: Tracking Any Point in Persistent 3D Geometrycited by 69
  6. 2024DELTA: Dense Efficient Long-range 3D Tracking for any videocited by 65
  7. 2021Weakly Supervised Learning of Rigid 3D Scene Flowcited by 60
  8. 2022Motion Inspired Unsupervised Perception and Prediction in Autonomous Drivingcited by 59
  9. 2021CamLiFlow: Bidirectional Camera-LiDAR Fusion for Joint Optical Flow and Scene Flow Estimationcited by 56
  10. 2021Scalable Scene Flow from Point Clouds in the Real Worldcited by 50
  11. 2021MultiBodySync: Multi-Body Segmentation and Motion Estimation via 3D Scan Synchronizationcited by 46
  12. 2023Fast Neural Scene Flowcited by 41
38 more
  1. 2023Hidden Gems: 4D Radar Scene Flow Learning Using Cross-Modal Supervisioncited by 36
  2. 2024ICP-Flow: LiDAR Scene Flow Estimation with ICPcited by 35
  3. 2021FESTA: Flow Estimation via Spatial-Temporal Attention for Scene Point Cloudscited by 30
  4. 2021Self-Supervised Multi-Frame Monocular Scene Flowcited by 29
  5. 2022Self-Supervised Scene Flow Estimation with 4-D Automotive Radarcited by 29
  6. 2024Recent Trends in 3D Reconstruction of General Non-Rigid Scenescited by 28
  7. 2022OcclusionFusion: Occlusion-aware Motion Estimation for Real-time Dynamic 3D Reconstructioncited by 27
  8. 2023Self-Supervised 3D Scene Flow Estimation Guided by Superpointscited by 24
  9. 2021SCTN: Sparse Convolution-Transformer Network for Scene Flow Estimationcited by 23
  10. 2023Instant-NVR: Instant Neural Volumetric Rendering for Human-object Interactions from Monocular RGBD Streamcited by 23
  11. 2023milliFlow: Scene Flow Estimation on mmWave Radar Point Cloud for Human Motion Sensingcited by 20
  12. 20223D Object Detection with a Self-supervised Lidar Scene Flow Backbonecited by 19
  13. 2025DGS-LRM: Real-Time Deformable 3D Gaussian Reconstruction From Monocular Videoscited by 15
  14. 2023GMSF: Global Matching Scene Flowcited by 14
  15. 2023Multi-Body Neural Scene Flowcited by 12
  16. 2024Match Stereo Videos via Bidirectional Alignmentcited by 11
  17. 2023Learning Optical Flow and Scene Flow with Bidirectional Camera-LiDAR Fusioncited by 10
  18. 2025DELTAv2: Accelerating Dense 3D Trackingcited by 10
  19. 2021Learning Scene Dynamics from Point Cloud Sequencescited by 9
  20. 2023Radar Instance Transformer: Reliable Moving Instance Segmentation in Sparse Radar Point Cloudscited by 9
  21. 2022RAFT-MSF: Self-Supervised Monocular Scene Flow using Recurrent Optimizercited by 7
  22. 2022Sequential Point Clouds: A Surveycited by 6
  23. 2024DiffSF: Diffusion Models for Scene Flow Estimationcited by 6
  24. 2024OptFlow: Fast Optimization-based Scene Flow Estimation without Supervisioncited by 5
  25. 2023EMR-MSF: Self-Supervised Recurrent Monocular Scene Flow Exploiting Ego-Motion Rigiditycited by 4
  26. 2021Temp-Frustum Net: 3D Object Detection with Temporal Fusioncited by 3
  27. 2022Learning Scene Flow in 3D Point Clouds with Noisy Pseudo Labelscited by 3
  28. 2023Attentive Multimodal Fusion for Optical and Scene Flowcited by 3
  29. 2024Self-Supervised Multi-Frame Neural Scene Flowcited by 3
  30. 2021Neural UpFlow: A Scene Flow Learning Approach to Increase the Apparent Resolution of Particle-Based Liquidscited by 2
  31. 2021MonoPLFlowNet: Permutohedral Lattice FlowNet for Real-Scale 3D Scene FlowEstimation with Monocular Imagescited by 2
  32. 2024TREND: Unsupervised 3D Representation Learning via Temporal Forecasting for LiDAR Perceptioncited by 2
  33. 2025Radar Velocity Transformer: Single-scan Moving Object Segmentation in Noisy Radar Point Cloudscited by 2
  34. 2023Exploiting Implicit Rigidity Constraints via Weight-Sharing Aggregation for Scene Flow Estimation from Point Cloudscited by 1
  35. 2024Fast Kernel Scene Flowcited by 1
  36. 2026x2-Fusion: Cross-Modality and Cross-Dimension Flow Estimation in Event Edge Spacecited by 1
  37. 2021D3DLO: Deep 3D LiDAR Odometrycited by 0
  38. 2023AGAR: Attention Graph-RNN for Adaptative Motion Prediction of Point Clouds of Deformable Objectscited by 0

50 most cited of 55 All 55 (JSON)