First open source realistic image generator: stable diffusion
GeneralThis forum is for non-technical general discussion which can include both Linux and non-Linux topics. Have fun!
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
First open source realistic image generator: stable diffusion
AI-powered image generators are not new, but so far they have all been proprietary. Stable Diffusion is open source. It has a special licence which prohibits certain obviously unethical uses, and there are some built-in security features such as watermarks, but it's surely only a matter of time before these are circumvented. At present photographic evidence has an important legal role. What will happen if anyone in the world can create any photrealistic image they want?
I don't think "image generator" is an accurate term, because it is not generating images, it is deriving them from other sources - including the copyrighted work of artists.
It is a similar problem to Microsoft's delicensing tool Copilot, which scans GitHub repos and offers derivatives of code found within, whilst expressly ignoring the license terms.
Humans can already create any photographic image they want - it just takes more time/effort/skill - but with digital camera images there's always the raw data from the camera's sensor - whilst I'm sure it is possible to forge that, I've not heard of any such attempts (presumably because the only reason for it would be fraud, and nobody is going to publicize doing that or make such tools widely available), so it just means photographic evidence may need to be examined a bit closer.
Still, I would be interested in the image blending functionality of Stable Diffusion, if it could be operated manually - i.e. if I chose which of my own images to feed in, especially if I could use masks/layers to define what got blended where - that would be cool.
Distribution: Arch Linux && OpenBSD 7.3 && Win 11 Pro
Posts: 736
Rep:
Quote:
Originally Posted by boughtonp
Still, I would be interested in the image blending functionality of Stable Diffusion, if it could be operated manually - i.e. if I chose which of my own images to feed in, especially if I could use masks/layers to define what got blended where - that would be cool.
yeah, thats what i have been thinkin also, making/drawing my own pictures and using those as a "seed".
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.