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

Is there two errors about comparative images in this paper? #43

Closed
angleboy8 opened this issue Apr 16, 2019 · 8 comments
Closed

Is there two errors about comparative images in this paper? #43

angleboy8 opened this issue Apr 16, 2019 · 8 comments

Comments

@angleboy8
Copy link

In Fig. 5, the last two images ocurs error. Others images are flipped up and down significantly.

@yulunzhang
Copy link
Owner

Hi
Fig5 has three groups of comparisons, could you point out which images exactly?

Also, all the results are reproduced by the public code and model. There’s no error in the results.

@angleboy8
Copy link
Author

angleboy8 commented Apr 16, 2019

Hi, the last two groups occur error. You carefully look other eight comparative images except for HR and RCAN one. Others images are flipped up and down significantly. Because of withtout attaching images, I only describe two errors.

@yulunzhang
Copy link
Owner

Hi,
There's no error. Let's take img_073 in Fig5 as an example. The results of other 8 methods are correct and the exact outputs by their codes and models.
You can run the SR results with other methods to see whether they would produce such results or not. Thanks.

@angleboy8
Copy link
Author

angleboy8 commented Apr 17, 2019

Please you pay more attention the texture direction of fig. Bicubic and fig. RCAN that aren't consistent with fig. HR in the last two groups. Obviously, test image and restoration image in this paper are wrong. If you need my help, I can send you a email that point out the detailed errors.

@yulunzhang
Copy link
Owner

Ok, please send me an email pointing out the error. yulun100@gmail.com
Thanks!

@angleboy8
Copy link
Author

Please notice your email. I just now sent a email. Thank you!

@angleboy8
Copy link
Author

Thank you for your reply. You're right. Indeed, other method exits some drawbacks that produce visual error for people's eyes, which results in my misunderstanding. Thanks a lot again. Best wishes!

@angleboy8
Copy link
Author

The latest version of Pytorch occurs the following error:

The "freeze_support()" line can be omitted if the program
is not going to be frozen to produce an executable.:

    An attempt has been made to start a new process before the
    current process has finished its bootstrapping phase.

    This probably means that you are not using fork to start your
    child processes and you have forgotten to use the proper idiom
    in the main module:

        if __name__ == '__main__':
            freeze_support()
            ...

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