SRMD-pytorch icon indicating copy to clipboard operation
SRMD-pytorch copied to clipboard

May you offer training code?

Open QueeneTam opened this issue 6 years ago • 7 comments

hi, I am a student who are doing a project about medical imageing, and I have a keen desire on your training code, hope that I can get your response from Email. Thanks a lot!

QueeneTam avatar Jul 25 '18 13:07 QueeneTam

@QueeneTam Of course, you can freely use the codes.

2wins avatar Jul 25 '18 13:07 2wins

ok,thanks a lot

QueeneTam avatar Jul 26 '18 08:07 QueeneTam

@2wins hi, I am sorry to bother you again, but I a deep-learning rookie did encounter some problems. I have been no idea what exactly the 'kernal' is. Is it a matrix of Guausse Blur, or anything else? I am preparing for my own medical dataset, but I don't understand how to make the kernal. Hope for your reply. Thanks a lot.

QueeneTam avatar Aug 13 '18 10:08 QueeneTam

@QueeneTam As you mentioned, Gaussian Blur kernel is a matrix, with which one can convolve an image to obtain a blurred image.

2wins avatar Aug 15 '18 04:08 2wins

@QueeneTam As you mentioned, Gaussian Blur kernel is a matrix, with which one can convolve an image to obtain a blurred image.

I am a student very interesting in your dimensionality stretching part implementation, could you please offer your training code or related code to dimensionality stretching in Email?

KindleHe avatar Sep 12 '18 14:09 KindleHe

@QueeneTam Of course, you can freely use the codes. hi, I am a student too who are doing a project about image resolution , and I have train a model follow your indicate, but the actual effect is very poor, i just put some images from DIV2K dataset to Dataset/train folder and train ,So could you please offer your trained model. I hope that I can get your response from Email. Thanks a lot!

xia00100 avatar Nov 12 '18 14:11 xia00100

Hi, could you teach me how to generate the degradation maps, and how to concatenate with the LR image? Could you upload the codes? Thanks very much!

hwuscut avatar Nov 14 '18 01:11 hwuscut