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

text prompt? #4

Open
hdjang opened this issue Oct 10, 2024 · 3 comments
Open

text prompt? #4

hdjang opened this issue Oct 10, 2024 · 3 comments

Comments

@hdjang
Copy link

hdjang commented Oct 10, 2024

Dear author
Thanks for sharing your great efforts!

I wonder how to get text prompt?
As I know, there is no text prompt provided in FFHQ original data.
How did you acquire those text prompts?

Best regards
HD Jang

@shivangi-aneja
Copy link
Owner

Hi,
We experimented with a fixed set of text prompts which we show in the paper. You are right, FFHQ does not come with text prompts. An example text prompt you can find in config/clipface.yaml in line 67-68.

@hdjang
Copy link
Author

hdjang commented Oct 14, 2024

Ah ha
Thanks for the reply!

How did you annotated those text prompt?

Did you get them by yourself via something like image-to-text model
or
did you use some data that already exists like below?
https://github.com/DCGM/ffhq-features-dataset

@shivangi-aneja
Copy link
Owner

Hi,
Since I tried experiments with only a small set of text prompts, I created them myself but you can also refer to other papers and see what other authors try out.
One example reference: https://github.com/orpatashnik/StyleCLIP

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