AI is going to be our biggest game changer
This discussion has been closed.
Daz 3D is part of 
Connect
DAZ Productions, Inc.
224 S 200 W, Suite #250
Salt Lake City, UT 84101
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
With Disney, using AI for the scripts might be real improvement
On the AI thing Adobe have just added in an AI option into Photoshop CC called Neural Filters, that adds some extra interesting tools for Postworking.
That is a compelling image. But with AI there always something that feels out of place, in this case the hand.
some goes with Stable Diffusion GUI on my own PC
text : Ernst Haeckel and various beings anatomy
This is what I was talking about in 2016 or so, I forget when sorry, in these forums, we've seen the text description generating AI art but that's not very flexible or directable really. What I was hoping for was that we coud write a really structured screenplay (think formal professional screenplay rules being shoehorned into an AI API / protocol) and then feed it to DAZ Studio to create our animation movie. I think it's a perfectly reasonable business product direction goal to develop an AI that can handle that type of work so that we have an Screenplay AI SDK that can be directly integrated into DS, Unity, UE, and so on.
Still have to see a single picture in this thread that wouldn't benefit from a hefty amount of postwork to remove some too obvious glitches in the Matrix.
Daz studio would have to be connect to a massive Data base
My local Data folder that supports my “modest” amount of Daz studio content, is nearly 30 gigabytes..
Imagine the amount of existing Data needed for your computer to parse a feature length animated movie from text prompts to say nothing of decisions about Character development exposition editorial direction and contextual dialog etc.
AI’s dont just “Imagine” new things it always needs existing reference material which is why” deep fake” videos are typical of highly photographed /filmed celebrities from the modern era .. instead of pre internet personalities with limited source material.
Not saying that this will never be possible but when it happens it will not be in a free program.
I remember like 2017 or 2018 Unity was working on one being able to feed their library of animations to a Unity API and Unity would then automatically analyze those animations and subdivide them into their component animations and create sub-animations from them. I don't know their progress as I've not check but human beings and other creautures are very limited actually in their range of motion. It seems like a job AI is ideally suited for. Likewise, a DAZ library like I have that takes up 1TB is actually very limited and something that AI should be able to analyze and categorize, even hiding leg, arms, and so on when poke through occurs. Although I think by the time a screenplay AI came about, the AI would be able to create characters, clothing, and such on the fly depending on how one directed it in the screenplay developed.
brought a tear to my eye
and I have been eating coleslaw all week too as lettuce too dear
I don't really know much what AI can and can't do currently, but here's my wishlist for Daz Studio AI
1) State of the art denoising/sharpening/resizing image manipulation AI. I think this is obvious. Huge time savings for everyone. If something like this could be hooked to render engine, you could render smaller noise free images fast and could later resize if needed
2) Fill the gaps AI. In animations you'd only need to render every 3rd or 4th frame or something and AI would fill in the rest. That would be a massive time saver and it should be possible to do at least when camera is stationary
3) AI which could correct poses/animations. It could check model's center of gravity, bend limits and stuff like that, and could assist users to get natural looking poses and movement. I wouldn't mind if that AI would also be environment aware, so that it could identify obstacles like in walk animations feet would not go through floor etc.
4) And of course an AI that could produce good text-to-speech, and had lipsync with Daz characters.
I agree but still would like to try and recreate it in 3D where I can fix things like that. I'm still reinstalling and just getting familiar with playing with stuff in DS and Zbrush but eventually I plan on trying to do a custom morph inspired by it and also the clothing. I think the dynamic long curly hair is a good match for what's in the picture. Plenty of good scenery that I can use...personally I'm not really interested in playing with ai. I seen some good images from friends who post on facebook. That said I enjoy the process of playing in zbrush and get a lot of satisfaction from doing creative things in 3d I don't think typing words into a dialogue would give me the same satisfaction. I'm also enjoying seeing the images that come up in ai. So far most are fairly derivative of known styles I'm curious as to whether it can come up with something truly original.
I agree, its the creative process that is most enjoyable part in the 3D art we create. Having AI do it, feels less.
Oh, it was cabbage based, not kale based after all... my bad.
both actually
Had the chance to start with OpenAI Dall-E this morning. I took its output for “wanted poster of futuristic female cyborg hanging on subway wall black and white illustration” and used it the image below. Hey-ho, I see a lot of promise in this new fangled AI art!
Cheers!
Sgt. Andrea: XNJ Crime Unit | Daz assets and set up | Blender Eevee render | Clp Studio Paint post-edit
I found another way to use it, DAZ renders as image prompts
Jack Tomalin's Danaides rendered with iray and some nearby buildings in Midjourney
That is where I tell folks, particularly artists, they need to start building their brand and IP. They are so adverse to working and collaborating with each other. 5 artists could have easily go together and in a span of 1 year created an animated movie and built their brand and characters and when AI comes out they just use their characters in a movie they generate with AI and do it in 3 months.
That is how I think artists thrive moving forward. Get with the base now and build and then use tthe tools to make it easier. But they never seem to listen.
I dunno... I forgot which one of the AI programs this was, but no matter what prompts I used to describe it, it couldn't accurately capture the essence of my inner child...
Okay the last one is a little accurate... but it's got a long way to go.
this looks cool
Those are eerie and cool. Well done.
Wendy, I think your stuff looks great! Please keep posting.
Yeah...it does. I've been having discussions on another forum where someone feels they could be used as backgrounds and possibly IBL's and that could be a valid use I suppose but I'm not sure I'd be willing to pay for ai backgrounds.
thannks
some cute Stable Diffusion GRisk GUI cats
( yes have lots of 4 legged one headed ones)
I'm sure everyone has heard this one by now... an AI generated artwork won a Colorado state fine art fair.
https://arstechnica.com/information-technology/2022/08/ai-wins-state-fair-art-contest-annoys-humans/?comments=1&post=41192494&mode=quote#reply
There is already a push by unionized artists in various fields to stop AI from taking their roles. This will only grow as AI works get more attention.
https://www.bbc.com/news/technology-61166272
This next link does not mention unions, but a recent news article that used an AI generated picture drew instant backlash from artists.
https://mixed-news.com/en/artists-and-designers-protest-against-ai-generated-graphics/
So I believe it is only a matter of time that their efforts will result in new copyright laws and usage restrictions that will curtail a lot of AI generated works from gaining too much ground. I think it will come down to how the AI was created. If you create an AI yourself using materials you have ownership of or permission to use, then you will be fine. But if you use one of these online AI generators in this thread, there will be no way to copyright your work.
The catch is obvious, not many people or companies are going to make their own AI. There will be exceptions, like I can see Disney creating an AI to replicate deceased actors, but even here they still need permission from the estate to use the likeness in the first place.
I know the Stable Diffusion says you own the results you get, but they also make it clear in the EULA that the laws of your country take priority:
"You agree not to use the Model or Derivatives of the Model:
- In any way that violates any applicable national, federal, state, local or international law or regulation"
Obviously this would include any copyright laws.
Additionally, the EULA for Stable sounds restrictive enough to limit naughty boys and girls from using it for their naughty purposes. So you naughty Daz artists out there have little to fear, LOL.
There will be all kinds of cool things you can do with this stuff, no doubt. But do not expect to be able to use them commercially so easily, and they will not replace real artists that easily, either, no more than a stock image seller has.
https://www.vice.com/en/article/bvmvqm/an-ai-generated-artwork-won-first-place-at-a-state-fair-fine-arts-competition-and-artists-are-pissed?
and some of mine
my biggest concern is not with AI art itself but the fallout for any digital art
lawmakers tend to sweep too far
to me full disclosure should be the thing required by law rather than outright banning
and laws about false representation of a person IE libel
a nonphotorealistic AI art rendering would fall under fair editorial use
apologies McGyver your post wasn't visible when I posted
There is also the possibility of people or companies secretly using AI and claiming they did the work themselves. That will certainly be an issue regardless of what laws get passed to restrict AI generated works. The bad actors out there will try to abuse it for their gain.
That is no different than any new technology, though. But the potential is pretty high for both good and bad to come from AI.
I think we'll truly be screwed when we see robots selling artificially generated velvet Elvis paintings out of 3D printed '78 Ford Econoline vans on the side of highways.
Once AI has mastered the velvet Elvis painting, all other art will become irrelevant.
All this AI work is derivative. Without being fed human-made images first, it would not generate a single piece. It's not that different than people making photo collages back in the day, except now it's automated.
It's going to fizzle because of the IP and copyrights. Someone owns those basic building blocks the AI is using, and the AI is only random generator incapable of making things from the scratch.