mlsanityData to 7 Oct 2026

No Pain, Big Gain: Classify Dynamic Point Cloud Sequences with Static Models by Fitting Feature-level Space-time Surfaces

arXiv 2203.11113 ↗First posted 21 Mar 2022cs.CV cs.AI cs.GRcited by 25

Not on the map. Why?

Abstract (arXiv)

Scene flow is a powerful tool for capturing the motion field of 3D point clouds. However, it is difficult to directly apply flow-based models to dynamic point cloud classification since the unstructured points make it hard or even impossible to efficiently and effectively trace point-wise correspondences. To capture 3D motions without explicitly tracking correspondences, we propose a kinematics-inspired neural network (Kinet) by generalizing the kinematic concept of ST-surfaces to the feature space. By unrolling the normal solver of ST-surfaces in the feature space, Kinet implicitly encodes feature-level dynamics and gains advantages from the use of mature backbones for static point cloud processing. With only minor changes in network structures and low computing overhead, it is painless to jointly train and deploy our framework with a given static model. Experiments on NvGesture, SHREC'17, MSRAction-3D, and NTU-RGBD demonstrate its efficacy in performance, efficiency in both the number of parameters and computational complexity, as well as its versatility to various static backbones. Noticeably, Kinet achieves the accuracy of 93.27% on MSRAction-3D with only 3.20M parameters and 10.35G FLOPS.

Key facts. Cited by 25 papers in the map (data through 7 Oct 2026). Builds on 77 of its 111 references that are in the map. First posted on arXiv 21 Mar 2022.

Builds on

77 of 111 references.

References · 77

  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. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,485
  4. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  5. 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,259
  6. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  7. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  8. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  9. 2018PointPillars: Fast Encoders for Object Detection from Point Cloudscited by 1,707
  10. 2017VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detectioncited by 1,323
  11. 2019KPConv: Flexible and Deformable Convolution for Point Cloudscited by 1,202
  12. 20194D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networkscited by 1,201
38 more
  1. 2018Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognitioncited by 1,199
  2. 2014Two-Stream Convolutional Networks for Action Recognition in Videoscited by 1,067
  3. 2018PointRCNN: 3D Object Proposal Generation and Detection from Point Cloudcited by 939
  4. 2016Temporal Segment Networks: Towards Good Practices for Deep Action Recognitioncited by 899
  5. 2015A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimationcited by 878
  6. 2018TSM: Temporal Shift Module for Efficient Video Understandingcited by 852
  7. 2018PointCNN: Convolution On X-Transformed Pointscited by 764
  8. 20173D Semantic Segmentation with Submanifold Sparse Convolutional Networkscited by 746
  9. 2016Multi-View 3D Object Detection Network for Autonomous Drivingcited by 721
  10. 2016NTU RGB+D: A Large Scale Dataset for 3D Human Activity Analysiscited by 683
  11. 2017Frustum PointNets for 3D Object Detection from RGB-D Datacited by 640
  12. 2019RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Cloudscited by 584
  13. 2015Multi-view Convolutional Neural Networks for 3D Shape Recognitioncited by 564
  14. 2019Deep Learning for 3D Point Clouds: A Surveycited by 557
  15. 2019NTU RGB+D 120: A Large-Scale Benchmark for 3D Human Activity Understandingcited by 530
  16. 2018PartNet: A Large-scale Benchmark for Fine-grained and Hierarchical Part-level 3D Object Understandingcited by 521
  17. 2016Convolutional Two-Stream Network Fusion for Video Action Recognitioncited by 384
  18. 2019Point-Voxel CNN for Efficient 3D Deep Learningcited by 343
  19. 2016OctNet: Learning Deep 3D Representations at High Resolutionscited by 340
  20. 2020Disentangling and Unifying Graph Convolutions for Skeleton-Based Action Recognitioncited by 339
  21. 2018Video Action Transformer Networkcited by 316
  22. 2018SpiderCNN: Deep Learning on Point Sets with Parameterized Convolutional Filterscited by 261
  23. 2018FlowNet3D: Learning Scene Flow in 3D Point Cloudscited by 241
  24. 2016Volumetric and Multi-View CNNs for Object Classification on 3D Datacited by 240
  25. 2019Actional-Structural Graph Convolutional Networks for Skeleton-based Action Recognitioncited by 228
  26. 2019Learning Object Bounding Boxes for 3D Instance Segmentation on Point Cloudscited by 156
  27. 2019An Attention Enhanced Graph Convolutional LSTM Network for Skeleton-Based Action Recognitioncited by 151
  28. 2021Offboard 3D Object Detection from Point Cloud Sequencescited by 144
  29. 2019STM: SpatioTemporal and Motion Encoding for Action Recognitioncited by 142
  30. 2020FLOT: Scene Flow on Point Clouds Guided by Optimal Transportcited by 129
  31. 20183DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentationcited by 118
  32. 2019MeteorNet: Deep Learning on Dynamic 3D Point Cloud Sequencescited by 117
  33. 2019HPLFlowNet: Hierarchical Permutohedral Lattice FlowNet for Scene Flow Estimation on Large-scale Point Cloudscited by 108
  34. 2019Just Go with the Flow: Self-Supervised Scene Flow Estimationcited by 98
  35. 2018View Adaptive Neural Networks for High Performance Skeleton-based Human Action Recognitioncited by 84
  36. 2020P2B: Point-to-Box Network for 3D Object Tracking in Point Cloudscited by 84
  37. 2017Optical Flow Guided Feature: A Fast and Robust Motion Representation for Video Action Recognitioncited by 82
  38. 2020Towards Semantic Segmentation of Urban-Scale 3D Point Clouds: A Dataset, Benchmarks and Challengescited by 80

50 most cited of 77 All 77 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 9; 2024: 3; 2025: 5; 2026: 7.

25 papers.

Sort:

By area:

  1. 2023PointCMP: Contrastive Mask Prediction for Self-supervised Learning on Point Cloud Videoscited by 22
  2. 2023Masked Spatio-Temporal Structure Prediction for Self-supervised Learning on Point Cloud Videoscited by 19
  3. 2026Align then Adapt: Rethinking Parameter-Efficient Transfer Learning in 4D Perceptioncited by 17
  4. 2023Point Contrastive Prediction with Semantic Clustering for Self-Supervised Learning on Point Cloud Videoscited by 16
  5. 2023X4D-SceneFormer: Enhanced Scene Understanding on 4D Point Cloud Videos through Cross-modal Knowledge Transfercited by 12
  6. 2024VG4D: Vision-Language Model Goes 4D Video Recognitioncited by 11
  7. 20233DInAction: Understanding Human Actions in 3D Point Cloudscited by 10
  8. 2023Multi-body SE(3) Equivariance for Unsupervised Rigid Segmentation and Motion Estimationcited by 7
  9. 2024Dynamic 3D Point Cloud Sequences as 2D Videoscited by 7
  10. 2022Meta-Sampler: Almost-Universal yet Task-Oriented Sampling for Point Cloudscited by 6
  11. 2023Contrastive Predictive Autoencoders for Dynamic Point Cloud Self-Supervised Learningcited by 5
  12. 2023NSM4D: Neural Scene Model Based Online 4D Point Cloud Sequence Understandingcited by 5
13 more
  1. 2026Tri-Efficient Transfer Learning for Point Cloud Videoscited by 5
  2. 2025Recognizing Actions from Robotic View for Natural Human-Robot Interactioncited by 4
  3. 2025Uni4D: A Unified Self-Supervised Learning Framework for Point Cloud Videoscited by 3
  4. 2024PRENet: A Plane-Fit Redundancy Encoding Point Cloud Sequence Network for Real-Time 3D Action Recognitioncited by 2
  5. 2025Content-Distortion High-Order Interaction for Blind Image Quality Assessmentcited by 2
  6. 2023A survey on deep learning approaches for data integration in autonomous driving systemcited by 1
  7. 2025DeSPITE: Exploring Contrastive Deep Skeleton-Pointcloud-IMU-Text Embeddings for Advanced Point Cloud Human Activity Understandingcited by 1
  8. 2025PointNet4D: A Lightweight 4D Point Cloud Video Backbone for Online and Offline Perception in Robotic Applicationscited by 1
  9. 2026SRENet: Spectral Re-Entry Network for Point Cloud Action Recognitioncited by 1
  10. 2026GATS: Gaussian Aware Temporal Scaling Transformer for Invariant 4D Spatio-Temporal Point Cloud Representationcited by 0
  11. 2026Paving the Way for Point Cloud Video Representation Learning Using A PDE Modelcited by 0
  12. 2026Joint-Embedding Prediction of Masked Point Tubes for Self-Supervised Learning on 4D Point Cloud Videoscited by 0
  13. 2026Motion-Saliency Complementary Masked Modeling for Point Cloud Video Understandingcited by 0