# PointPWC-Net: A Coarse-to-Fine Network for Supervised and Self-Supervised Scene Flow Estimation on 3D Point Clouds

arXiv: https://arxiv.org/abs/1911.12408 · HTML: https://mlsanity.com/paper/1911.12408/

## Key facts

- Cited by 38 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Nov 2019.

## Built on by (most cited first, 30 of 38)

- [4DComplete: Non-Rigid Motion Estimation Beyond the Observable Surface](https://mlsanity.com/paper/2105.01905/) (2021)
- [Lepard: Learning partial point cloud matching in rigid and deformable scenes](https://mlsanity.com/paper/2111.12591/) (2021)
- [Rotation-Invariant Transformer for Point Cloud Matching](https://mlsanity.com/paper/2303.08231/) (2023)
- [Weakly Supervised Learning of Rigid 3D Scene Flow](https://mlsanity.com/paper/2102.08945/) (2021)
- [CamLiFlow: Bidirectional Camera-LiDAR Fusion for Joint Optical Flow and Scene Flow Estimation](https://mlsanity.com/paper/2111.10502/) (2021)
- [Scalable Scene Flow from Point Clouds in the Real World](https://mlsanity.com/paper/2103.01306/) (2021)
- [MultiBodySync: Multi-Body Segmentation and Motion Estimation via 3D Scan Synchronization](https://mlsanity.com/paper/2101.06605/) (2021)
- [Non-rigid Point Cloud Registration with Neural Deformation Pyramid](https://mlsanity.com/paper/2205.12796/) (2022)
- [Accurate Point Cloud Registration with Robust Optimal Transport](https://mlsanity.com/paper/2111.00648/) (2021)
- [FESTA: Flow Estimation via Spatial-Temporal Attention for Scene Point Clouds](https://mlsanity.com/paper/2104.00798/) (2021)
- [Efficient 3D Deep LiDAR Odometry](https://mlsanity.com/paper/2111.02135/) (2021)
- [Zero-Shot Monocular Scene Flow Estimation in the Wild](https://mlsanity.com/paper/2501.10357/) (2025)
- [3D Object Detection with a Self-supervised Lidar Scene Flow Backbone](https://mlsanity.com/paper/2205.00705/) (2022)
- [GMSF: Global Matching Scene Flow](https://mlsanity.com/paper/2305.17432/) (2023)
- [DELFlow: Dense Efficient Learning of Scene Flow for Large-Scale Point Clouds](https://mlsanity.com/paper/2308.04383/) (2023)
- [Learning Optical Flow and Scene Flow with Bidirectional Camera-LiDAR Fusion](https://mlsanity.com/paper/2303.12017/) (2023)
- [Floxels: Fast Unsupervised Voxel Based Scene Flow Estimation](https://mlsanity.com/paper/2503.04718/) (2025)
- [Attention Models for Point Clouds in Deep Learning: A Survey](https://mlsanity.com/paper/2102.10788/) (2021)
- [Sequential Point Clouds: A Survey](https://mlsanity.com/paper/2204.09337/) (2022)
- [Self-Supervised Bird's Eye View Motion Prediction with Cross-Modality Signals](https://mlsanity.com/paper/2401.11499/) (2024)
- [Dynamic Semantic Occupancy Mapping using 3D Scene Flow and Closed-Form Bayesian Inference](https://mlsanity.com/paper/2108.03180/) (2021)
- [Occlusion-aware Non-Rigid Point Cloud Registration via Unsupervised Neural Deformation Correntropy](https://mlsanity.com/paper/2502.10704/) (2025)
- [ContrastMotion: Self-supervised Scene Motion Learning for Large-Scale LiDAR Point Clouds](https://mlsanity.com/paper/2304.12589/) (2023)
- [Diff-Reg v1: Diffusion Matching Model for Registration Problem](https://mlsanity.com/paper/2403.19919/) (2024)
- [A Unified Framework for Human-centric Point Cloud Video Understanding](https://mlsanity.com/paper/2403.20031/) (2024)
- [DetFlowTrack: 3D Multi-object Tracking based on Simultaneous Optimization of Object Detection and Scene Flow Estimation](https://mlsanity.com/paper/2203.02157/) (2022)
- [Slice Transformer and Self-supervised Learning for 6DoF Localization in 3D Point Cloud Maps](https://mlsanity.com/paper/2301.08957/) (2023)
- [Unleash the Potential of 3D Point Cloud Modeling with A Calibrated Local Geometry-driven Distance Metric](https://mlsanity.com/paper/2306.00552/) (2023)
- [HumanReg: Self-supervised Non-rigid Registration of Human Point Cloud](https://mlsanity.com/paper/2312.05462/) (2023)
- [SSFlowNet: Semi-supervised Scene Flow Estimation On Point Clouds With Pseudo Label](https://mlsanity.com/paper/2312.15271/) (2023)

## Abstract (arXiv)

We propose a novel end-to-end deep scene flow model, called PointPWC-Net, on 3D point clouds in a coarse-to-fine fashion. Flow computed at the coarse level is upsampled and warped to a finer level, enabling the algorithm to accommodate for large motion without a prohibitive search space. We introduce novel cost volume, upsampling, and warping layers to efficiently handle 3D point cloud data. Unlike traditional cost volumes that require exhaustively computing all the cost values on a high-dimensional grid, our point-based formulation discretizes the cost volume onto input 3D points, and a PointConv operation efficiently computes convolutions on the cost volume. Experiment results on FlyingThings3D outperform the state-of-the-art by a large margin. We further explore novel self-supervised losses to train our model and achieve comparable results to state-of-the-art trained with supervised loss. Without any fine-tuning, our method also shows great generalization ability on KITTI Scene Flow 2015 dataset, outperforming all previous methods.