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

Missing images after preprocessing FFHQ's in-the-wild-images #70

Closed
thuanz123 opened this issue Sep 25, 2022 · 4 comments
Closed

Missing images after preprocessing FFHQ's in-the-wild-images #70

thuanz123 opened this issue Sep 25, 2022 · 4 comments

Comments

@thuanz123
Copy link

I have download the original in-the-wild-images for preprocessing and have validate that all original 70000 images is not corrupted, however after preprocessing using the instruction in the README, the final crop folder miss 43 images. Is this expected or your code is not correct ?

@oneThousand1000
Copy link

I think those images can not be detected by mtcnn, so the authors removed them from eg3d training dataset.

@thuanz123
Copy link
Author

I think those images can not be detected by mtcnn, so the authors removed them from eg3d training dataset.

So is it OK to train eg3d with these images missing ?

@oneThousand1000
Copy link

I think those images can not be detected by mtcnn, so the authors removed them from eg3d training dataset.

So is it OK to train eg3d with these images missing ?

Yes, it is ok. In #30 (comment) someone got nice results using only ~5k images.

@thuanz123
Copy link
Author

Thanks a lot. I will close this then

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