This repository contains the codes and models for the paper "Motion Deblurring with Real Events" [Paper] [Supplementary] [Project Page]

If you use the code, models or data for your research, please cite us accordingly:

@inproceedings{xu2021motion,
  title={Motion Deblurring with Real Events},
  author={Xu, Fang and Yu, Lei and Wang, Bishan and Yang, Wen and Xia, Gui-Song and Jia, Xu and Qiao, Zhendong and Liu, Jianzhuang},
  booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},
  pages={2583--2592},
  year={2021}
}

You can download the pretrained RED-RBE model from here

This code has been tested with CUDA 11.0 and Python 3.6.

pip install -r requirements.txt

You can download a sample sequence from this link, which is converted from [HQF].

Use the following command to test the neural network:

python test_deblur.py

This code is based on the code available in the slow-motion repo. I am grateful to the authors for making the original source code available.

We are glad to hear if you have any suggestions and questions.

Please send email to [email protected]