Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
167 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

ICP-Flow: LiDAR Scene Flow Estimation with ICP (2402.17351v2)

Published 27 Feb 2024 in cs.CV

Abstract: Scene flow characterizes the 3D motion between two LiDAR scans captured by an autonomous vehicle at nearby timesteps. Prevalent methods consider scene flow as point-wise unconstrained flow vectors that can be learned by either large-scale training beforehand or time-consuming optimization at inference. However, these methods do not take into account that objects in autonomous driving often move rigidly. We incorporate this rigid-motion assumption into our design, where the goal is to associate objects over scans and then estimate the locally rigid transformations. We propose ICP-Flow, a learning-free flow estimator. The core of our design is the conventional Iterative Closest Point (ICP) algorithm, which aligns the objects over time and outputs the corresponding rigid transformations. Crucially, to aid ICP, we propose a histogram-based initialization that discovers the most likely translation, thus providing a good starting point for ICP. The complete scene flow is then recovered from the rigid transformations. We outperform state-of-the-art baselines, including supervised models, on the Waymo dataset and perform competitively on Argoverse-v2 and nuScenes. Further, we train a feedforward neural network, supervised by the pseudo labels from our model, and achieve top performance among all models capable of real-time inference. We validate the advantage of our model on scene flow estimation with longer temporal gaps, up to 0.4 seconds where other models fail to deliver meaningful results.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (61)
  1. Slim: Self-supervised lidar scene flow and motion segmentation. In International Conference on Computer Vision (ICCV), 2021a.
  2. Slim: Self-supervised lidar scene flow and motion segmentation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 13126–13136, 2021b.
  3. Pointflownet: Learning representations for rigid motion estimation from point clouds. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 7962–7971, 2019.
  4. Method for registration of 3-d shapes. In Sensor fusion IV: control paradigms and data structures, pages 586–606. Spie, 1992.
  5. nuscenes: A multimodal dataset for autonomous driving. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11621–11631, 2020.
  6. Density-based clustering based on hierarchical density estimates. In Pacific-Asia conference on knowledge discovery and data mining, pages 160–172. Springer, 2013.
  7. Object modelling by registration of multiple range images. Image and vision computing, 10(3):145–155, 1992.
  8. Bi-pointflownet: Bidirectional learning for point cloud based scene flow estimation. In European Conference on Computer Vision, pages 108–124. Springer, 2022.
  9. Re-evaluating lidar scene flow for autonomous driving. arXiv preprint arXiv:2304.02150, 2023.
  10. Deep global registration. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 2514–2523, 2020.
  11. David F Crouse. On implementing 2d rectangular assignment algorithms. IEEE Transactions on Aerospace and Electronic Systems, 52(4):1679–1696, 2016.
  12. Rigid scene flow for 3d lidar scans. In 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 1765–1770. IEEE, 2016.
  13. Exploiting rigidity constraints for lidar scene flow estimation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 12776–12785, 2022.
  14. 3d object detection with a self-supervised lidar scene flow backbone. In European Conference on Computer Vision, pages 247–265. Springer, 2022.
  15. Weakly supervised learning of rigid 3d scene flow. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5692–5703, 2021.
  16. Hplflownet: Hierarchical permutohedral lattice flownet for scene flow estimation on large-scale point clouds. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 3254–3263, 2019.
  17. Dynamic 3d scene analysis by point cloud accumulation. In European Conference on Computer Vision, pages 674–690. Springer, 2022.
  18. Deformation and correspondence aware unsupervised synthetic-to-real scene flow estimation for point clouds. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 7233–7243, 2022.
  19. Scalable scene flow from point clouds in the real world. IEEE Robotics and Automation Letters, 7(2):1589–1596, 2021.
  20. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014.
  21. Flowstep3d: Model unrolling for self-supervised scene flow estimation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 4114–4123, 2021.
  22. Scoop: Self-supervised correspondence and optimization-based scene flow. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 5281–5290, 2023.
  23. Patchwork++: Fast and robust ground segmentation solving partial under-segmentation using 3d point cloud. In 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 13276–13283. IEEE, 2022.
  24. Self-point-flow: Self-supervised scene flow estimation from point clouds with optimal transport and random walk. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 15577–15586, 2021a.
  25. Rigidflow: Self-supervised scene flow learning on point clouds by local rigidity prior. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 16959–16968, 2022.
  26. Neural scene flow prior. Advances in Neural Information Processing Systems, 34:7838–7851, 2021b.
  27. Fast neural scene flow. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pages 9878–9890, 2023.
  28. Flownet3d: Learning scene flow in 3d point clouds. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 529–537, 2019a.
  29. Meteornet: Deep learning on dynamic 3d point cloud sequences. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 9246–9255, 2019b.
  30. Accelerated hierarchical density based clustering. In Data Mining Workshops (ICDMW), 2017 IEEE International Conference on, pages 33–42. IEEE, 2017.
  31. hdbscan: Hierarchical density based clustering. The Journal of Open Source Software, 2(11):205, 2017.
  32. Object scene flow for autonomous vehicles. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 3061–3070, 2015.
  33. Just go with the flow: Self-supervised scene flow estimation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11177–11185, 2020.
  34. Motion inspired unsupervised perception and prediction in autonomous driving. In European Conference on Computer Vision, pages 424–443. Springer, 2022.
  35. Colored point cloud registration revisited. In Proceedings of the IEEE international conference on computer vision, pages 143–152, 2017.
  36. FLOT: Scene Flow on Point Clouds Guided by Optimal Transport. In European Conference on Computer Vision, 2020.
  37. Accelerating 3d deep learning with pytorch3d. arXiv:2007.08501, 2020.
  38. Caspr: Learning canonical spatiotemporal point cloud representations. Advances in neural information processing systems, 33:13688–13701, 2020.
  39. Efficient variants of the icp algorithm. In Proceedings third international conference on 3-D digital imaging and modeling, pages 145–152. IEEE, 2001.
  40. Fast point feature histograms (fpfh) for 3d registration. In 2009 IEEE international conference on robotics and automation, pages 3212–3217. IEEE, 2009.
  41. Scalability in perception for autonomous driving: Waymo open dataset. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 2446–2454, 2020.
  42. Self-supervised learning of non-rigid residual flow and ego-motion. In 2020 international conference on 3D vision (3DV), pages 150–159. IEEE, 2020.
  43. A learning approach for real-time temporal scene flow estimation from lidar data. In 2017 IEEE International Conference on Robotics and Automation (ICRA), pages 5666–5673. IEEE, 2017.
  44. Zeroflow: Fast zero label scene flow via distillation. arXiv preprint arXiv:2305.10424, 2023.
  45. Multi-body neural scene flow. arXiv preprint arXiv:2310.10301, 2023.
  46. Kiss-icp: In defense of point-to-point icp–simple, accurate, and robust registration if done the right way. IEEE Robotics and Automation Letters, 8(2):1029–1036, 2023.
  47. 3d scene flow estimation with a rigid motion prior. In 2011 International Conference on Computer Vision, pages 1291–1298. IEEE, 2011.
  48. Piecewise rigid scene flow. In Proceedings of the IEEE International Conference on Computer Vision, pages 1377–1384, 2013.
  49. Festa: Flow estimation via spatial-temporal attention for scene point clouds. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 14173–14182, 2021.
  50. Deep closest point: Learning representations for point cloud registration. In Proceedings of the IEEE/CVF international conference on computer vision, pages 3523–3532, 2019.
  51. 4d unsupervised object discovery. Advances in Neural Information Processing Systems, 35:35563–35575, 2022.
  52. Flownet3d++: Geometric losses for deep scene flow estimation. In Proceedings of the IEEE/CVF winter conference on applications of computer vision, pages 91–98, 2020.
  53. Stereoscopic scene flow computation for 3d motion understanding. International Journal of Computer Vision, 95:29–51, 2011.
  54. 3D Multi-Object Tracking: A Baseline and New Evaluation Metrics. IROS, 2020a.
  55. AB3DMOT: A Baseline for 3D Multi-Object Tracking and New Evaluation Metrics. ECCVW, 2020b.
  56. Argoverse 2: Next generation datasets for self-driving perception and forecasting. In Proceedings of the Neural Information Processing Systems Track on Datasets and Benchmarks (NeurIPS Datasets and Benchmarks 2021), 2021.
  57. Pointpwc-net: Cost volume on point clouds for (self-) supervised scene flow estimation. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part V 16, pages 88–107. Springer, 2020.
  58. Teaser: Fast and certifiable point cloud registration. IEEE Transactions on Robotics, 37(2):314–333, 2020.
  59. Flowmot: 3d multi-object tracking by scene flow association. arXiv preprint arXiv:2012.07541, 2020.
  60. 3d registration with maximal cliques. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 17745–17754, 2023.
  61. Open3d: A modern library for 3d data processing. arXiv preprint arXiv:1801.09847, 2018.
Citations (8)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
Youtube Logo Streamline Icon: https://streamlinehq.com

YouTube