mlsanityData to 7 Oct 2026

Occlusion Guided Self-supervised Scene Flow Estimation on 3D Point Clouds

arXiv 2104.04724 ↗First posted 10 Apr 2021cs.CVcited by 11

Not on the map. Why?

Abstract (arXiv)

Understanding the flow in 3D space of sparsely sampled points between two consecutive time frames is the core stone of modern geometric-driven systems such as VR/AR, Robotics, and Autonomous driving. The lack of real, non-simulated, labeled data for this task emphasizes the importance of self- or un-supervised deep architectures. This work presents a new self-supervised training method and an architecture for the 3D scene flow estimation under occlusions. Here we show that smart multi-layer fusion between flow prediction and occlusion detection outperforms traditional architectures by a large margin for occluded and non-occluded scenarios. We report state-of-the-art results on Flyingthings3D and KITTI datasets for both the supervised and self-supervised training.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Builds on 30 of its 53 references that are in the map. First posted on arXiv 10 Apr 2021.

Builds on

30 of 53 references.

References · 30

  1. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  2. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  3. 2015FlowNet: Learning Optical Flow with Convolutional Networkscited by 1,054
  4. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 878
  5. 2018PointConv: Deep Convolutional Networks on 3D Point Cloudscited by 656
  6. 2019Point-Voxel CNN for Efficient 3D Deep Learningcited by 343
  7. 2018FlowNet3D: Learning Scene Flow in 3D Point Cloudscited by 241
  8. 2016Volumetric and Multi-View CNNs for Object Classification on 3D Datacited by 240
  9. 2017UnFlow: Unsupervised Learning of Optical Flow with a Bidirectional Census Losscited by 202
  10. 2021Deep Parametric Continuous Convolutional Neural Networkscited by 161
  11. 2019Iterative Residual Refinement for Joint Optical Flow and Occlusion Estimationcited by 131
  12. 2020FLOT: Scene Flow on Point Clouds Guided by Optimal Transportcited by 129
18 more
  1. 2019SelFlow: Self-Supervised Learning of Optical Flowcited by 128
  2. 2020Learning by Analogy: Reliable Supervision from Transformations for Unsupervised Optical Flow Estimationcited by 120
  3. 2020MaskFlownet: Asymmetric Feature Matching with Learnable Occlusion Maskcited by 117
  4. 2019HPLFlowNet: Hierarchical Permutohedral Lattice FlowNet for Scene Flow Estimation on Large-scale Point Cloudscited by 108
  5. 2017Occlusion Aware Unsupervised Learning of Optical Flowcited by 101
  6. 2019Just Go with the Flow: Self-Supervised Scene Flow Estimationcited by 98
  7. 2018Models Matter, So Does Training: An Empirical Study of CNNs for Optical Flow Estimationcited by 94
  8. 2020FlowStep3D: Model Unrolling for Self-Supervised Scene Flow Estimationcited by 71
  9. 2018Occlusions, Motion and Depth Boundaries with a Generic Network for Disparity, Optical Flow or Scene Flow Estimationcited by 69
  10. 2020Self-Supervised Monocular Scene Flow Estimationcited by 64
  11. 2019DDFlow: Learning Optical Flow with Unlabeled Data Distillationcited by 63
  12. 2020Self-Supervised Learning of Non-Rigid Residual Flow and Ego-Motioncited by 41
  13. 2020DeepLiDARFlow: A Deep Learning Architecture For Scene Flow Estimation Using Monocular Camera and Sparse LiDARcited by 24
  14. 2017MirrorFlow: Exploiting Symmetries in Joint Optical Flow and Occlusion Estimationcited by 21
  15. 2019PWOC-3D: Deep Occlusion-Aware End-to-End Scene Flow Estimationcited by 20
  16. 2020Occlusion Guided Scene Flow Estimation on 3D Point Cloudscited by 17
  17. 2017Multiframe Scene Flow with Piecewise Rigid Motioncited by 6
  18. 2020Adversarial Self-Supervised Scene Flow Estimationcited by 6

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 3; 2023: 4; 2024: 2; 2025: 1.

11 papers.

  1. 2022Dynamic 3D Scene Analysis by Point Cloud Accumulationcited by 47
  2. 2021Multiway Non-rigid Point Cloud Registration via Learned Functional Map Synchronizationcited by 30
  3. 2022SCOOP: Self-Supervised Correspondence and Optimization-Based Scene Flowcited by 28
  4. 2023Self-Supervised 3D Scene Flow Estimation Guided by Superpointscited by 24
  5. 2023Self-supervised Learning for Pre-Training 3D Point Clouds: A Surveycited by 21
  6. 2022Unsupervised Learning of 3D Scene Flow with 3D Odometry Assistancecited by 2
  7. 2023Self-Supervised 3D Scene Flow Estimation and Motion Prediction using Local Rigidity Priorcited by 2
  8. 2025Weakly and Self-Supervised Class-Agnostic Motion Prediction for Autonomous Drivingcited by 2
  9. 2023SSFlowNet: Semi-supervised Scene Flow Estimation On Point Clouds With Pseudo Labelcited by 1
  10. 2024RMS-FlowNet++: Efficient and Robust Multi-Scale Scene Flow Estimation for Large-Scale Point Cloudscited by 1
  11. 2024SSRFlow: Semantic-aware Fusion with Spatial Temporal Re-embedding for Real-world Scene Flowcited by 0