kin-zhang/OpenSceneFlow

Scene Flow Models for Autonomous Driving Dataset

Please check the usage on KTH-RPL/DeFlow or KTH-RPL/SeFlow.

• DeFlow: Supervised learning scene flow, model included is trained on Argoverse 2
• SeFlow: Self-Supervised learning scene flow, model included is trained on Argoverse 2. Paper also reported Waymo result, the weight cannot be shared according to Waymo Term. More detail discussion issue 8.

The files we included and all test result reports can be found v2 leaderboard and v1 leaderboard.

• seflow_best.ckpt: trained on Argoverse 2 sensor set with self-supervised way.
• deflow_best.ckpt: trained on Argoverse 2 sensor set with ground truth using deflowLoss function.
• fastflow3d_best.ckpt: trained on Argoverse 2 sensor set with ground truth using ff3dLoss function.
• ... more models on the way
• demo_data.zip: 613Mb, a mini-dataset for user to quickly run train/val code. Check usage in this section.
• waymo_map.tar.gz: to successfully process waymo data with ground segmentation included to unified h5 file. Check usage in this README.

🎁 One repository, All methods!

You can try following methods in our code without any effort to make your own benchmark.

• SeFlow (Ours 🚀): ECCV 2024
• DeFlow (Ours 🚀): ICRA 2024
• FastFlow3d: RA-L 2021
• ZeroFlow: ICLR 2024, their pre-trained weight can covert into our format easily through the script.
• NSFP: NeurIPS 2021, faster 3x than original version because of our CUDA speed up, same (slightly better) performance. Done coding, public after review.
• FastNSF: ICCV 2023. Done coding, public after review.
• ... more on the way

Citation

@inproceedings{zhang2024deflow,
  author={Zhang, Qingwen and Yang, Yi and Fang, Heng and Geng, Ruoyu and Jensfelt, Patric},
  booktitle={2024 IEEE International Conference on Robotics and Automation (ICRA)}, 
  title={{DeFlow}: Decoder of Scene Flow Network in Autonomous Driving}, 
  year={2024},
  pages={2105-2111},
  doi={10.1109/ICRA57147.2024.10610278}
}
@inproceedings{zhang2024seflow,
  author={Zhang, Qingwen and Yang, Yi and Li, Peizheng and Andersson, Olov and Jensfelt, Patric},
  title={{SeFlow}: A Self-Supervised Scene Flow Method in Autonomous Driving},
  booktitle={European Conference on Computer Vision (ECCV)},
  year={2024},
  pages={353–369},
  organization={Springer},
  doi={10.1007/978-3-031-73232-4_20},
}