Ryuichi @ryuichi@qoto.org

@nop i think he said somehwere that the project wasn't dead, he just was updating DEV instead of MASTER

#ai #stablediffusion

@nop We need to determine if AI can process and interpret actions that it sees through video feeds, and respond accordingly. Once we witness robots like Boston Dynamics' Atlas equipped with AI, reacting to their environment in real time, we'll know that the time has come.

We're currently experiencing a moment of change that may be comparable only to the industrial revolution. It's both scary to consider the possibilities but also i feel fortunate to live in such a transformative period.

#ai #stablediffusion

@nop In the case of characters, maybe you can ask the AI for rough bases using simple ideas and the artists can refine and expand those ideas, it would leverage work from a lot of people in the anime production field, but i don't think it can replace artists ever, you still need people to animate and do a proper draw of characters in all angles, you just can't do that consistently with AI.

#ai #stablediffusion #dailyrei

I'm not sure how you feel about AI-generated images, I think it's causing a stir in the art world like we had never seen before. I like to generate images but I don't consider myself someone who's trying to steal the spotlight from artists since I'm not creating anything. I still have great respect and admiration for artists who have honed their skills through learning and practice.

But one positive aspect of AI-generated images is that they allow for a free flow of ideas and concepts that might have otherwise remained hidden. People who lack the skill or time to express themselves can now do so with greater clarity thanks to AI-generated images.

Rather than seeing AI-generated images and traditional art as competitors, I believe that they can be complementary. AI-generated images can serve as a useful tool for artists to overcome creative and technical barriers and improve their skills. Meanwhile, those who appreciate and generate images with the help of AI can help push art into a new era by exploring new ideas in the graphic space.

#ai #stablediffusion

#ai #stablediffusion

@maid i remember when i started generating images with Anything V3, that model i think needed more explicit prompts to be able to get what you wanted, but nowadays these new models are more smart and you get better stuff with less tags.

#ai #stablediffusion

#ai #stablediffusion #dailyrei

@nop specialized Language AIs trained on specific data will surely become a thing in the following years

@nop related to hands i just have "extra hand, extra hands" and i think it helped a bit, but related to fingers i don't have anything, i just accept whatever it comes.

Some people that release models do recommend stuff like the "easynegative" lora/textual but i've used it and i don't like it's influence.

I don't know if someone has done it, but maybe the way of doing a good negative would be to make a LORA trained on BAD images and then use it as a negative, perhaps.