jixiang2016 / rss-t Goto Github PK

Paper | Supp | Dataset

Xiang Ji¹, Zhixiang Wang^1,2, Shin'ichi Satoh^2,1, Yinqiang Zheng¹

¹The University of Tokyo  ²National Institute of Informatics  

This repository provides the official PyTorch implementation of the paper.

This paper explores a novel in-between exposure mode called global reset release (GRR) shutter, which produces GS-like blur but with row-dependent blur magnitude. We take advantage of this unique characteristic of GRR to explore the latent frames within a single image and restore a clear counterpart by relying only on these latent contexts.

1. Python and Pytorch

• Pyhotn=3.8 (Anaconda recommended)
• Pytorch=1.11.0
• CUDA=11.3/11.4

conda create -n rsst python=3.8
conda activate rsst
conda install pytorch==1.11.0 torchvision==0.12.0 torchaudio==0.11.0 cudatoolkit=11.3 -c pytorch

2. Other packages

pip install -r requirements.txt

• Download datasets GRR_real and RSGR-GS_v1.
• Unzip them under a specified directory by yourself.
• The dataset folder structure should be like the format below (Minor adjustments to the folder structure of RSGR-GS_v1 may be needed.):

GRR_real
├─ train
│ ├─ seq1    % 50 sequences
│ │ ├─ GS
| | | ├─ PNG
| | | | ├─xxxx.png
| | | | ├─......
| | | ├─ RAW (same as PNG)
| | | | ├─ ......
│ │ ├─ RSGR (same as GS)
| | | ├─ ......
│ │
│─ validate  %  7 sequences
│ ├─ ...... (same as train)
│
├─ test    % 7 sequences
│ ├─ ...... (same as train)

• Please download checkpoints from this link and put them under root directory of this project.

To test RSS-T, please run the command below:

bash ./test.sh       ### Please specify your data directory and output path in the script

To train RSS-T, please run the command below:

bash ./train.sh       ### Please refer to the script for more info.

If you find our work useful, please kindly cite as:

@InProceedings{Ji_2023_ICCV,
    author    = {Ji, Xiang and Wang, Zhixiang and Satoh, Shin'ichi and Zheng, Yinqiang},
    title     = {Single Image Deblurring with Row-dependent Blur Magnitude},
    booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
    month     = {October},
    year      = {2023},
    pages     = {12269-12280}
}

This project is based on public works below:

• neural-global-shutter
• MIMO-UNet
• Uformer
• DeepDeblur_release