Home

Awesome

Joint Demosaicing and Deghosting of Time-Varying Exposures for Single-Shot HDR Imaging

Project page | Paper | Supplemental

Jungwoo Kim, Min H. Kim KAIST

Tested Environments

Installation

This repository is built in Pytorch 1.9.0 and tested on Ubuntu 16.04 enviornment (Python3.7, CUDA10.2, cuDNN7.6).

Please follow below instructions:

  1. Clone our repository
git clone git@github.com:KAIST-VCLAB/singshot-hdr-demosaicing.git
cd singshot-hdr-demosaicing
  1. Make conda enviornment
conda create -n pytorch190 python=3.7
conda activate pytorch190
  1. Install dependencies
conda install pytorch==1.9.0 torchvision==0.10.0 torchaudio==0.9.0 cudatoolkit=10.2 -c pytorch
pip install -r "requirements.txt"

Demo

First download our pretrained model and put best_psnr_mu.pt file in code/models/best_psnr_mu.pt.

To run demo with pre-trained models, run below code:

cd code
python test.py