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.

Sign up for GitHub

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

Jump to bottom

Question in paper: 1. how to get the mask region as input; 2. Any comparison with ProPainter, just curious #1

Open
LokiXun opened this issue Oct 12, 2023 · 1 comment
Open

Question in paper: 1. how to get the mask region as input; 2. Any comparison with ProPainter, just curious #1

LokiXun opened this issue Oct 12, 2023 · 1 comment

Comments

@LokiXun
Copy link

LokiXun commented Oct 12, 2023 •
edited
Loading

For Q1. how to get the mask region as input; Maybe it just doing frame-wise decrease with GT and binarize the mask?
@LIUTIGHE
Copy link
Owner

LIUTIGHE commented Oct 12, 2023 •
edited
Loading

  1. The mask region in our dataset is extracted by processing difference map between the original frame and the corrupted frame (i.e. morphorlogical, binarize, etc.), which will be introduced in detail in our supplementary materials. In the real application, it requires users to provide frame-wise masks.
  2. As of the submission and rebuttal period of our paper, propainter has not released their code, so our paper does not include comparison results with it. But recently we have noticed that influential work and conducted tests. The preliminary test results indicate that propainter can improve, but cannot "revolutionarily" improve the performance of the inpainting method. We believe that one of the main reasons is still the unrecoverable flow in our task setting.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@LokiXun @LIUTIGHE