Generalizable-Non-Line-of-Sight-Imaging-with-Learnable-Physical-Priors

This is the official codebase of the paper Generalizable-Non-Line-of-Sight-Imaging-with-Learnable-Physical-Priors (ICCV2025).

Dataset

Synthetic Dataset

We utilized the synthetic data (~3000 motorbike dataset) provided by LFE.

Please download in Here.

Real-world Data

We utilized the real-world data provided by FK and NLOST.

Besides, the self-captured data can be found at Here.

Training

1. Download the synthetic dataset.
2. Modify the data path and result root in DL_inference/inference/config.py.
3. Run train_256.py.

Acknowledgements

We thank the authors who shared the code of their works. Particularly LFE.

Citation

@inproceedings{sun2025generalizable,
  title={Generalizable Non-Line-of-Sight Imaging with Learnable Physical Priors},
  author={Sun, Shida and Li, Yue and Zhang, Yueyi and Xiong, Zhiwei},
  booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},
  year={2025}
}