milliFlow: Scene Flow Estimation on mmWave Radar Point Cloud for Human Motion Sensing

This work presents milliFlow, a scene flow estimation module to provide an additional layer of point-wise motion information on top of the original mmWave radar point cloud in the conventional mmWave-based human motion sensing pipeline. For technical details, please refer to our paper on ECCV 2024:

milliFlow: Scene Flow Estimation on mmWave Radar Point Cloud for Human Motion Sensing
Fangqiang Ding, Zhen Luo, Peijun Zhao, Chris Xiaoxuan Lu
[arXiv] [video] [poster]

News

• [2024-03-15] Our preprint paper is available on 👉arXiv.
• [2024-07-01] Our paper is accepted by 🎉ECCV 2024.
• [2024-09-12] Our presentation video and poster is online. Please check them out 👉video | poster

Data

You can download the dataset here.

Getting Started

To find out how to run our scodes, please see our intructions in GETTING_STARTED, which will be made by the end of October. Stay tuned for update!

Citation

When using the code, model, or dataset, please cite the following paper:

@InProceedings{Ding_2024_ECCV,
    author    = {Ding, Fangqiang and Luo, Zhen and Zhao, Peijun and Lu, Chris Xiaoxuan},
    title     = {milliFlow: Scene Flow Estimation on mmWave Radar Point Cloud for Human Motion Sensing},
    booktitle = {Proceedings of the European Conference on Computer Vision  (ECCV)},
    year      = {2024},
    pages     = {1-14}
}

License

Code and Model License

The code and model provided in this repository are licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). See the LICENSE file for details.

Dataset License

The dataset provided in this repository is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). See the LICENSE file for details.