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

How to specify custom transfer mask? #30

Open
netsalesbg-ltd opened this issue May 23, 2023 · 3 comments
Open

How to specify custom transfer mask? #30

netsalesbg-ltd opened this issue May 23, 2023 · 3 comments

Comments

@netsalesbg-ltd
Copy link

In the readme you said you we can use custom transfer mask, but couldn't find any additional info about how to do that? So how can I specify custom transfer mask?

@BadourAlBahar
Copy link
Owner

You can think of it as some sort of copy paste. The mask will tell you the region. You can draw the mask your self of use human parsing to specify the garment masks and then utilize this to select your region of interest.

@netsalesbg-ltd
Copy link
Author

netsalesbg-ltd commented May 23, 2023

I get that, I have already setup a human parser, I get the garment, but I am having trouble replacing part_mask in GarmentTransferSpatialAppearanceEncoder. Cannot understand what format it should be, for the rest of the garment transfer script to be able to use it.

Edit, to clarify. This is the result I currently get.
image

Since the model copies based on the torso from densepose, if the source image has a short top, part of the bottom is also copied over to the target. Now, I am using human parsing to separate the top correctly.
image

But cannot figure out how to implement it in the code, I cannot understand the result of GarmentTransferSpatialAppearanceEncoder. Please can you explain the output of GarmentTransferSpatialAppearanceEncoder and how I can use the mask from human parsing in the Generator. Thank you!

@i-amgeek
Copy link

@netsalesbg-ltd Hi, have you been able to solve it?

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

3 participants