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

[FEAT] Generate face based off of a PNG of a real-life person #55

Open
a4v2d4 opened this issue Feb 8, 2025 · 7 comments
Open

[FEAT] Generate face based off of a PNG of a real-life person #55

a4v2d4 opened this issue Feb 8, 2025 · 7 comments

Comments

@a4v2d4
Copy link

a4v2d4 commented Feb 8, 2025

This would essentially be like Vectorizer.ai, but the SVG would follow your current style. So you could easily create face svg's of actual prof athletes without infringing on whatever licenses or whatever are stopping common folk like us from using.

I was thinking of using some modern AI image analysis stuff (even if simply asking chatgpt for feature values manually) to analyze player headshot pngs/jpegs and output values for all face 'types' values... Then that could be used with CLI to generate actual SVG's with facejs styling?

If anyone can think of a better way, let me know!! I might just take a stab at something like this myself.

@dumbmatter
Copy link
Member

So you could easily create face svg's of actual prof athletes without infringing on whatever licenses or whatever are stopping common folk like us from using.

I'm not sure if that is legally true, even if you own the rights to the original image.

I was thinking of using some modern AI image analysis stuff (even if simply asking chatgpt for feature values manually) to analyze player headshot pngs/jpegs and output values for all face 'types' values... Then that could be used with CLI to generate actual SVG's with facejs styling?

It's an interesting idea. Would be really cool if you could get it to work. People would have fun putting themselves in it.

I'm not sure how easy it will be to do. Can an LLM do it out of the box, maybe with a little instruction? Maybe? How good can they do it? I don't know? Would be cool to see what you come up with!

@a4v2d4
Copy link
Author

a4v2d4 commented Feb 9, 2025

I'm not sure if that is legally true, even if you own the rights to the original image.

Could you explain/expand please? Even if I own the rights to the image? If I drew a picture of my favorite player by hand, what do you think the legal restrictions would be for how I use it?

@dumbmatter
Copy link
Member

Publicity rights - can't necessarily use pictures of real people (even a cartoon you made) for all purposes. Depends on the details though.

@a4v2d4
Copy link
Author

a4v2d4 commented Feb 9, 2025

Fudge, I'm dumb. Forgot what NIL stood for 😂

@a4v2d4
Copy link
Author

a4v2d4 commented Feb 9, 2025

@dumbmatter is there an easy place that lists all the different valid values for each feature type?

Like, I see the feature types defined here: https://github.com/zengm-games/facesjs/blob/master/src/types.ts.

And I can figure out the valid numeric ranges for scalar values from the random generations here: https://github.com/zengm-games/facesjs/blob/master/src/generate.ts.

But idk where, without clicking through web ui, i can find all the values for the different ID types (e.g. hat2 or jersey1)? Am I missing something?

@tomkennedy22
Copy link
Contributor

You can find each of the individual SVGs in the /svg folder under their component subfolders

And then the best consolidated list is probably here - https://github.com/zengm-games/facesjs/blob/master/tools/lib/genders.js

@a4v2d4
Copy link
Author

a4v2d4 commented Feb 9, 2025

@tomkennedy22 Thank you!

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