mlsanityData to 7 Oct 2026

FESTA: Flow Estimation via Spatial-Temporal Attention for Scene Point Clouds

arXiv 2104.00798 ↗First posted 1 Apr 2021cs.CVcited by 30

Not on the map. Why?

Abstract (arXiv)

Scene flow depicts the dynamics of a 3D scene, which is critical for various applications such as autonomous driving, robot navigation, AR/VR, etc. Conventionally, scene flow is estimated from dense/regular RGB video frames. With the development of depth-sensing technologies, precise 3D measurements are available via point clouds which have sparked new research in 3D scene flow. Nevertheless, it remains challenging to extract scene flow from point clouds due to the sparsity and irregularity in typical point cloud sampling patterns. One major issue related to irregular sampling is identified as the randomness during point set abstraction/feature extraction -- an elementary process in many flow estimation scenarios. A novel Spatial Abstraction with Attention (SA^2) layer is accordingly proposed to alleviate the unstable abstraction problem. Moreover, a Temporal Abstraction with Attention (TA^2) layer is proposed to rectify attention in temporal domain, leading to benefits with motions scaled in a larger range. Extensive analysis and experiments verified the motivation and significant performance gains of our method, dubbed as Flow Estimation via Spatial-Temporal Attention (FESTA), when compared to several state-of-the-art benchmarks of scene flow estimation.

Key facts. Cited by 30 papers in the map (data through 7 Oct 2026). Builds on 22 of its 37 references that are in the map. First posted on arXiv 1 Apr 2021.

Builds on

22 of 37 references.

References · 22

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  3. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  4. 2017Non-local Neural Networkscited by 2,290
  5. 2015FlowNet: Learning Optical Flow with Convolutional Networkscited by 1,054
  6. 2016A Point Set Generation Network for 3D Object Reconstruction from a Single Imagecited by 972
  7. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 878
  8. 2016FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networkscited by 780
  9. 2019Deep Learning for 3D Point Clouds: A Surveycited by 557
  10. 2018Pyramid Stereo Matching Networkcited by 486
  11. 2018FlowNet3D: Learning Scene Flow in 3D Point Cloudscited by 241
  12. 2019Modeling Point Clouds with Self-Attention and Gumbel Subset Samplingcited by 151
10 more
  1. 2019MeteorNet: Deep Learning on Dynamic 3D Point Cloud Sequencescited by 117
  2. 2019HPLFlowNet: Hierarchical Permutohedral Lattice FlowNet for Scene Flow Estimation on Large-scale Point Cloudscited by 108
  3. 2019Just Go with the Flow: Self-Supervised Scene Flow Estimationcited by 98
  4. 2019SampleNet: Differentiable Point Cloud Samplingcited by 74
  5. 2018Occlusions, Motion and Depth Boundaries with a Generic Network for Disparity, Optical Flow or Scene Flow Estimationcited by 69
  6. 2019FlowNet3D++: Geometric Losses For Deep Scene Flow Estimationcited by 55
  7. 2019Adaptive Hierarchical Down-Sampling for Point Cloud Classificationcited by 40
  8. 2019PointPWC-Net: A Coarse-to-Fine Network for Supervised and Self-Supervised Scene Flow Estimation on 3D Point Cloudscited by 38
  9. 2015ABCNN: Attention-Based Convolutional Neural Network for Modeling Sentence Pairscited by 35
  10. 2018Zoom and Learn: Generalizing Deep Stereo Matching to Novel Domainscited by 20

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 1; 2022: 4; 2023: 10; 2024: 9; 2025: 2; 2026: 4.

30 papers.

Sort:

By area:

  1. 2023Fast Neural Scene Flowcited by 39
  2. 2023Hidden Gems: 4D Radar Scene Flow Learning Using Cross-Modal Supervisioncited by 36
  3. 2024ICP-Flow: LiDAR Scene Flow Estimation with ICPcited by 34
  4. 2022What Matters for 3D Scene Flow Networkcited by 31
  5. 2022SCOOP: Self-Supervised Correspondence and Optimization-Based Scene Flowcited by 28
  6. 2023Self-Supervised 3D Scene Flow Estimation Guided by Superpointscited by 24
  7. 2021SCTN: Sparse Convolution-Transformer Network for Scene Flow Estimationcited by 23
  8. 2023milliFlow: Scene Flow Estimation on mmWave Radar Point Cloud for Human Motion Sensingcited by 20
  9. 2025Zero-Shot Monocular Scene Flow Estimation in the Wildcited by 20
  10. 2026Align then Adapt: Rethinking Parameter-Efficient Transfer Learning in 4D Perceptioncited by 17
  11. 2023GMSF: Global Matching Scene Flowcited by 14
  12. 2023DELFlow: Dense Efficient Learning of Scene Flow for Large-Scale Point Cloudscited by 14
18 more
  1. 2023Multi-Body Neural Scene Flowcited by 12
  2. 2023DifFlow3D: Toward Robust Uncertainty-Aware Scene Flow Estimation with Diffusion Modelcited by 11
  3. 2024DiffSF: Diffusion Models for Scene Flow Estimationcited by 6
  4. 2022Sequential Point Clouds: A Surveycited by 5
  5. 2024OptFlow: Fast Optimization-based Scene Flow Estimation without Supervisioncited by 5
  6. 2022Learning Scene Flow in 3D Point Clouds with Noisy Pseudo Labelscited by 3
  7. 2024Self-Supervised Multi-Frame Neural Scene Flowcited by 3
  8. 2023Exploiting Implicit Rigidity Constraints via Weight-Sharing Aggregation for Scene Flow Estimation from Point Cloudscited by 1
  9. 2024Fast Kernel Scene Flowcited by 1
  10. 2024CMU-Flownet: Exploring Point Cloud Scene Flow Estimation in Occluded Scenariocited by 1
  11. 2024RMS-FlowNet++: Efficient and Robust Multi-Scale Scene Flow Estimation for Large-Scale Point Cloudscited by 1
  12. 2023AGAR: Attention Graph-RNN for Adaptative Motion Prediction of Point Clouds of Deformable Objectscited by 0
  13. 2024STARFlow: Spatial Temporal Feature Re-embedding with Attentive Learning for Real-world Scene Flowcited by 0
  14. 2024SSRFlow: Semantic-aware Fusion with Spatial Temporal Re-embedding for Real-world Scene Flowcited by 0
  15. 2025VISC: mmWave Radar Scene Flow Estimation using Pervasive Visual-Inertial Supervisioncited by 0
  16. 2026SF3D-RGB: Scene Flow Estimation from Monocular Camera and Sparse LiDARcited by 0
  17. 2026PCSTracker: Long-Term Scene Flow Estimation for Point Cloud Sequencescited by 0
  18. 2026RVLoss: Runoff Vote Loss for Self-Supervised LiDAR Scene Flow Estimationcited by 0