|
1 | 1 | DeltaFlow |
2 | 2 | --- |
3 | 3 |
|
4 | | -Status: Under Review 🚀 |
| 4 | +[](https://arxiv.org/abs/2508.17054) |
| 5 | +<!-- [](https://paperswithcode.com/sota/self-supervised-scene-flow-estimation-on-1?p=seflow-a-self-supervised-scene-flow-method-in) --> |
| 6 | +<!-- [](https://hkustconnect-my.sharepoint.com/:b:/g/personal/qzhangcb_connect_ust_hk/EWyWD-tAX4xIma5U7ZQVk9cBVjsFv0Y_jAC2G7xAB-w4cg?e=c3FbMg) --> |
| 7 | +<!-- [](https://youtu.be/fQqx2IES-VI) --> |
5 | 8 |
|
6 | | -We appreciate your interest! The work is actively under peer review. |
| 9 | +Status: Under Review 🚀 |
7 | 10 |
|
8 | | -We will release the full codebase, trained models, supplementary materials and all baselines upon acceptance and publication. It may around early October 2025. |
| 11 | +We will release the full codebase, trained models, supplementary materials and all baselines upon acceptance and publication. |
9 | 12 |
|
10 | 13 | Stay tuned and feel free to star ⭐ this repository to get notified when we publish. |
11 | 14 |
|
12 | 15 |
|
13 | 16 | 🔗 While you wait, explore our codebase: [KTH-RPL/OpenSceneFlow](https://github.com/KTH-RPL/OpenSceneFlow). |
14 | 17 |
|
15 | | -- 2025/08/24: I'm updating some codes for early release. Please be patient. |
16 | | -- 2025/08/24: Updating train data augmentation as illustrated in the DeltaFlow paper. |
| 18 | +**News w. TBD**: |
| 19 | + |
| 20 | +- 2025/08/24: I'm updating some codes for early release. |
| 21 | +- [x] 2025/08/24: Updating train data augmentation as illustrated in the DeltaFlow paper. |
| 22 | +- [x] 2025/08/25: Updating paper preprint link. |
| 23 | +- [ ] DeltaFlow Model python file. |
| 24 | +- [ ] pre-trained weights upload. |
| 25 | +- [ ] DeltaFlow Loss fn. |
| 26 | +- [ ] Merged into [OpenSceneFlow](https://github.com/KTH-RPL/OpenSceneFlow) |
| 27 | + |
| 28 | +## Cite & Acknowledgements |
| 29 | + |
| 30 | +``` |
| 31 | +@article{zhang2025deltaflow, |
| 32 | + title={{DeltaFlow}: An Efficient Multi-frame Scene Flow Estimation Method}, |
| 33 | + author={Zhang, Qingwen and Zhu, Xiaomeng and Zhang, Yushan and Cai, Yixi and Andersson, Olov and Jensfelt, Patric}, |
| 34 | + year={2025}, |
| 35 | + journal={arXiv preprint arXiv:2508.17054}, |
| 36 | +} |
| 37 | +``` |
17 | 38 |
|
| 39 | +This work was partially supported by the Wallenberg AI, Autonomous Systems and Software Program (WASP) funded by the Knut and Alice Wallenberg Foundation and Prosense (2020-02963) funded by Vinnova. |
0 commit comments