Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Without down sampling the test image #96

Open
abdulhanan24 opened this issue Aug 24, 2019 · 3 comments
Open

Without down sampling the test image #96

abdulhanan24 opened this issue Aug 24, 2019 · 3 comments

Comments

@abdulhanan24
Copy link

abdulhanan24 commented Aug 24, 2019

This evalute.py actually take the input image and then simply down sample the image first and then apply bicubic and then apply dcscn which almost similar to the input image. I don't want to down sample the image so for example if you give 2 x 2 resolution image as test image it would return 4 x 4.
Please reply thanks

@jiny2001
Copy link
Owner

Hi,
In that case, I think you can just use sr.py. evaluate.py is for checking super resolution performance. Thx!

@abdulhanan24
Copy link
Author

Thanks its worked.
If we have multiple options from High resolution to low resolution for same dataset, What you suggest on which resolution should we train so system can generate better results on all resolution

@jiny2001
Copy link
Owner

It's totally up to what scale you want. Please note if you trained the model with x3, the model only work for x3 super resolution. There is a technic to share x2 - x4 model for the first part and it worth try.
However, so far, my model only works for one specific scale. Thx!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants