CameraSR
CameraSR copied to clipboard
issue in preprocessing train data
hi,@ngchc, thanks for you release your great work. I have one problem with the preprocessing train data.
what is this mean? and why do this step ?
one more issue. Is train data raw or png?
Hi. For the first question, since there exists the intensity variation between captured LR and HR images, we perform a compensation process to address this issue. The mean value (i.e., the DC component in the frequence domain) is adopted to calculate the strength of this compensation. For the second one, both of them are provided. Please refer to the README file for detailed information.
Hi. For the first question, since there exists the intensity variation between captured LR and HR images, we perform a compensation process to address this issue. The mean value (i.e., the DC component in the frequence domain) is adopted to calculate the strength of this compensation. For the second one, both of them are provided. Please refer to the README file for detailed information.
thanks for good job ,could you release your data processing code?
@ngchc thanks. I have the same question with @llp1996 , could you release your data processing code?
@yanmenglu @llp1996 Thanks for your interest. We don't have a plan to release it for the time being. The provided datasets and models are sufficient to reproduce our work.