# Re-Evaluating LiDAR Scene Flow for Autonomous Driving

arXiv: https://arxiv.org/abs/2304.02150 · HTML: https://mlsanity.com/paper/2304.02150/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Builds on 43 of its 85 references that are in the map.
- First posted on arXiv 4 Apr 2023.

## Builds on (most cited first, 30 of 43)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation](https://mlsanity.com/paper/1612.00593/) (2016)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space](https://mlsanity.com/paper/1706.02413/) (2017)
- [DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation](https://mlsanity.com/paper/1901.05103/) (2019)
- [Implicit Neural Representations with Periodic Activation Functions](https://mlsanity.com/paper/2006.09661/) (2020)
- [Occupancy Networks: Learning 3D Reconstruction in Function Space](https://mlsanity.com/paper/1812.03828/) (2018)
- [PointPillars: Fast Encoders for Object Detection from Point Clouds](https://mlsanity.com/paper/1812.05784/) (2018)
- [Learning Implicit Fields for Generative Shape Modeling](https://mlsanity.com/paper/1812.02822/) (2018)
- [A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation](https://mlsanity.com/paper/1512.02134/) (2015)
- [PointCNN: Convolution On X-Transformed Points](https://mlsanity.com/paper/1801.07791/) (2018)
- [Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecasting](https://mlsanity.com/paper/2301.00493/) (2023)
- [SAL: Sign Agnostic Learning of Shapes from Raw Data](https://mlsanity.com/paper/1911.10414/) (2019)
- [FlowNet3D: Learning Scene Flow in 3D Point Clouds](https://mlsanity.com/paper/1806.01411/) (2018)
- [KISS-ICP: In Defense of Point-to-Point ICP -- Simple, Accurate, and Robust Registration If Done the Right Way](https://mlsanity.com/paper/2209.15397/) (2022)
- [FLOT: Scene Flow on Point Clouds Guided by Optimal Transport](https://mlsanity.com/paper/2007.11142/) (2020)
- [RAFT-3D: Scene Flow using Rigid-Motion Embeddings](https://mlsanity.com/paper/2012.00726/) (2020)
- [MeteorNet: Deep Learning on Dynamic 3D Point Cloud Sequences](https://mlsanity.com/paper/1910.09165/) (2019)
- [HPLFlowNet: Hierarchical Permutohedral Lattice FlowNet for Scene Flow Estimation on Large-scale Point Clouds](https://mlsanity.com/paper/1906.05332/) (2019)
- [Just Go with the Flow: Self-Supervised Scene Flow Estimation](https://mlsanity.com/paper/1912.00497/) (2019)
- [Neural Scene Flow Prior](https://mlsanity.com/paper/2111.01253/) (2021)
- [PV-RAFT: Point-Voxel Correlation Fields for Scene Flow Estimation of Point Clouds](https://mlsanity.com/paper/2012.00987/) (2020)
- [Smooth Shells: Multi-Scale Shape Registration with Functional Maps](https://mlsanity.com/paper/1905.12512/) (2019)
- [FlowStep3D: Model Unrolling for Self-Supervised Scene Flow Estimation](https://mlsanity.com/paper/2011.10147/) (2020)
- [Self-Supervised Monocular Scene Flow Estimation](https://mlsanity.com/paper/2004.04143/) (2020)
- [Weakly Supervised Learning of Rigid 3D Scene Flow](https://mlsanity.com/paper/2102.08945/) (2021)
- [Motion Inspired Unsupervised Perception and Prediction in Autonomous Driving](https://mlsanity.com/paper/2210.08061/) (2022)
- [Deep Rigid Instance Scene Flow](https://mlsanity.com/paper/1904.08913/) (2019)
- [Scalable Scene Flow from Point Clouds in the Real World](https://mlsanity.com/paper/2103.01306/) (2021)
- [Patchwork++: Fast and Robust Ground Segmentation Solving Partial Under-Segmentation Using 3D Point Cloud](https://mlsanity.com/paper/2207.11919/) (2022)

## Built on by (most cited first, 17 of 17)

- [SeFlow: A Self-Supervised Scene Flow Method in Autonomous Driving](https://mlsanity.com/paper/2407.01702/) (2024)
- [Any4D: Unified Feed-Forward Metric 4D Reconstruction](https://mlsanity.com/paper/2512.10935/) (2025)
- [milliFlow: Scene Flow Estimation on mmWave Radar Point Cloud for Human Motion Sensing](https://mlsanity.com/paper/2306.17010/) (2023)
- [SSF: Sparse Long-Range Scene Flow for Autonomous Driving](https://mlsanity.com/paper/2501.17821/) (2025)
- [Multi-Body Neural Scene Flow](https://mlsanity.com/paper/2310.10301/) (2023)
- [HiMo: High-Speed Objects Motion Compensation in Point Clouds](https://mlsanity.com/paper/2503.00803/) (2025)
- [DeltaFlow: An Efficient Multi-frame Scene Flow Estimation Method](https://mlsanity.com/paper/2508.17054/) (2025)
- [TeFlow: Enabling Multi-frame Supervision for Self-Supervised Feed-forward Scene Flow Estimation](https://mlsanity.com/paper/2602.19053/) (2026)
- [DoGFlow: Self-Supervised LiDAR Scene Flow via Cross-Modal Doppler Guidance](https://mlsanity.com/paper/2508.18506/) (2025)
- [Self-Supervised Multi-Frame Neural Scene Flow](https://mlsanity.com/paper/2403.16116/) (2024)
- [Let-It-Flow: Simultaneous Optimization of 3D Flow and Object Clustering](https://mlsanity.com/paper/2404.08363/) (2024)
- [TerraSeg: Self-Supervised Ground Segmentation for Any LiDAR](https://mlsanity.com/paper/2603.27344/) (2026)
- [SynFlow: Scaling Up LiDAR Scene Flow Estimation with Synthetic Data](https://mlsanity.com/paper/2604.09411/) (2026)
- [Fast Kernel Scene Flow](https://mlsanity.com/paper/2403.05896/) (2024)
- [UniFlow: Zero-Shot LiDAR Scene Flow for Autonomous Vehicles](https://mlsanity.com/paper/2511.18254/) (2025)
- [Weakly Supervised Cross-Modal Learning for 4D Radar Scene Flow Estimation](https://mlsanity.com/paper/2605.18507/) (2026)
- [CorrelationFlow: A Training-Free Geometric Approach for LiDAR Scene Flow Estimation](https://mlsanity.com/paper/2607.29237/) (2026)

## Abstract (arXiv)

Popular benchmarks for self-supervised LiDAR scene flow (stereoKITTI, and FlyingThings3D) have unrealistic rates of dynamic motion, unrealistic correspondences, and unrealistic sampling patterns. As a result, progress on these benchmarks is misleading and may cause researchers to focus on the wrong problems. We evaluate a suite of top methods on a suite of real-world datasets (Argoverse 2.0, Waymo, and NuScenes) and report several conclusions. First, we find that performance on stereoKITTI is negatively correlated with performance on real-world data. Second, we find that one of this task's key components -- removing the dominant ego-motion -- is better solved by classic ICP than any tested method. Finally, we show that despite the emphasis placed on learning, most performance gains are caused by pre- and post-processing steps: piecewise-rigid refinement and ground removal. We demonstrate this through a baseline method that combines these processing steps with a learning-free test-time flow optimization. This baseline outperforms every evaluated method.