Code repo for the paper "DAVANet: Stereo Deblurring with View Aggregation" (CVPR'19, Oral).  [Paper]   [Project Page] 

Download the dataset (192.5GB, unzipped 202.2GB) from [Data Website].

You could download the pretrained model (34.8MB) of DAVANet from [Here].

(Note that the model does not need to unzip, just load it directly.)

• Linux (tested on Ubuntu 14.04/16.04)
• Python 2.7+
• Pytorch 0.4.1
• easydict
• tensorboardX
• pyexr

pip install -r requirements.txt

Use the following command to train the neural network:

python runner.py 
        --phase 'train'\
        --data [dataset path]\
        --out [output path]

Use the following command to test the neural network:

python runner.py \
        --phase 'test'\
        --weights './ckpt/best-ckpt.pth.tar'\
        --data [dataset path]\
        --out [output path]

Use the following command to resume training the neural network:

python runner.py 
        --phase 'resume'\
        --weights './ckpt/best-ckpt.pth.tar'\
        --data [dataset path]\
        --out [output path]

You can also use the following simple command, with changing the settings in config.py:

python runner.py

If you find DAVANet, or Stereo Blur Dataset useful in your research, please consider citing:

@inproceedings{zhou2019davanet,
  title={{DAVANet}: Stereo Deblurring with View Aggregation},
  author={Zhou, Shangchen and Zhang, Jiawei and Zuo, Wangmeng and Xie, Haozhe and Pan, Jinshan and Ren, Jimmy},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  year={2019}
}

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

Please send email to [email protected]

[1] Zhe Hu, Li Xu, and Ming-Hsuan Yang. Joint depth estimation and camera shake removal from single blurry image. In CVPR, 2014.

[2] Seungjun Nah, Tae Hyun Kim, and Kyoung Mu Lee. Deep multi-scale convolutional neural network for dynamic scene deblurring. In CVPR, 2017.

[3] Orest Kupyn, Volodymyr Budzan, Mykola Mykhailych, Dmytro Mishkin, and Jiri Matas. Deblurgan: Blind motion deblurring using conditional adversarial networks. In CVPR, 2018.

[4] Jiawei Zhang, Jinshan Pan, Jimmy Ren, Yibing Song, Lin- chao Bao, Rynson WH Lau, and Ming-Hsuan Yang. Dynamic scene deblurring using spatially variant recurrent neural networks. In CVPR, 2018.

[5] Xin Tao, Hongyun Gao, Xiaoyong Shen, Jue Wang, and Jiaya Jia. Scale-recurrent network for deep image deblurring. In CVPR, 2018.

This project is open sourced under MIT license.