AI is going to be our biggest game changer
This discussion has been closed.
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2025 Daz Productions Inc. All Rights Reserved.
Comments
LOl. I am living in one of the most interesting eras.
Now, we have Daz Studio social media ADs being bombed by Anti-AI commenters and Daz users arguing about what Daz is.
A user said it's 3D models and a rendering engine and someone else said that they were lying and Daz was another AI art app.
-And even that Daz Studio was another scam-app and THEN, then someone said Daz was expensive.
I hope they meant that 'buying assets' could grow to an expense, but it read like the software was expensive.
And people kept saying IT'S FREE, IT'S FREE!
But still being ignored.
No, I said it doesn't work by make X% new and it's no longer a potential infringment.
It's an interesting topic, though, where DAZ will be at some point in the future. Some of the seemingly odd points will probably become actual questions, like "Can it still be free?", or "Will it marry an ai?".
finally, the "make art" button exists
a lot of ignorant people who cannot draw the curtains are jumping onto digital artists of all types accusing them of using AI it seems.
a big proportion of the anti AI crowd are not artists either, I know a few on social media myself.
Of course lots of artists are against it but a lot are fully embracing it too and many that are are Fine Artists using paint and other mediums.
They are using it to create references
Most digital artists can pick the difference.
Meh, opinions are like bumholes... You know the rest...
So far, AI works are having a little trouble getting their copyright recognized.
Copyright Revoked For AI Comic After Office Learns How AI Works (kotaku.com)
Now is that an 'educated guess'?
This video shows how good AI is getting. The creator of this video used ChatGPT, DALL-E2 and Elevenlabs Voice AI to create the story images and voices.
Sorry, I look at that and think - this tech has a long way to go before it will convince me.
I feel you are doing what humans do... failing to appreciate non-linear phenomena. To better see this, consider that none of this was even remotely possible, say, two years ago. Now imagine a software system that is better than today's, by a wider degree than today's is better than two years ago.
Also, objectively, that video is garbage - So much of the ai generated nonsense is garbage, and we're burning the earth more to achieve it.
In context, it's a statement of fact. Since I do know the rest, I agree with N-Rarts' statement.
I was responding to the previous observation. I watched the video and, in my opinion, it sucks regardless of the super-advanced tech.
However, in general, I'm totally unimpressed by all the AI hype. No doubt it will find its niche in our technological toy-box but human creativity is of a different order, again IMHO.
the photo editing possibilities jusy keep improving exponentially
That does look very cool, even what Adobe are adding into Photoshop with their new Neural Filters option is starting to look very good.
Great, so that's your opinion?
It's a pretty general statement for a random place on the internet...
This is a specific place on the internet, and it applies here as much as in any other place.
Oh - NOT with cool outpainting, or any other A.I. way! This was OLD SKOOL Photoshop.
I generated the image in MidJourney, then fired up DAZ Studio (who am I kidding, it is ALWAYS up and running!) and tried to eyeball the pose and camera angle, matched the light as best I could and generated the render. Bring the MJ image into Photoshop, bring the DAZ image in as a layer over the top and mask everything out, carefully bringing in just the eyes. For what it is worth the eyes are from Herschel Hoffmeyer's Belle.
Same technique could as well be used to composite in a DAZ rendered face to get hair, environment and clothing that MJ generated. Everything is just one tool in a toolbox!
Latest Corridor video:
And the behind the scenes:
Both the Corridor videos are so interesting. Thanks for sharing.
we can use our DAZ characters and costumes to train embeddings of consistent characters
the embedding I trained on my face does consistently reproduce my face
I just also have the flickering issue, ControlNet is helping to add more coherence, things are moving very fast
Indeed. I slightly twitch on the diffusion part, because it's based on the "learn to reproduce original image/image as close as possible part", so "like human" may apply in a way, but likely without the breadths and depths of the real-time processing (minus a couple of millions/billions of neurons in play), imagine "reproduce x as closely as possible vs. what would i make of this '?'". Numbers don't always count, though, and abilities for visualizing (image/audio/whatever/math/life/universe) vary wildy amongst humans. Many professionals from "different" fields hail/"expect officially" on new ai, often medium-obviously wrongly, simply. "Up to" the best of the (currently) best people on some field, like stock markets. Certainly not entirely wrong, but also not quite right. Not meaning to interfere with people using and trying out stuff out there. Just a mild reminder for those, considering rather overall long-term-progress questions.
This video provide some instructions.
Make stunning AI Animation from your videos
... and one more ...
The Key to Unlocking Better AI Animations!
I am curious why all those YouTubers like using gif2gif as opposed to batch image conversion
I can do 30fps 1minute 1800 frame videos quite easily using image batches
applying the same seed, parameters, prompt etc to all
coherence depending on noise and cfg, (my crazier ones were so because of settings not the batch process)
(depending on dimensions can take a while but no difference to what he is doing )
saves a lot of time not doing all those conversions to gif and back
Forgive a newbie question that may factually be just wishful thinking, but..... Has anyone yet figured out a way to extract a character from an AI platform (Midjourney, for instance) into something that can be used in DAZ?
Or Blender?
Or Poser?
Or am I a few years too early?
Things are moving so fast you may only be a few MINUTES too early. I have seen some tests where a 3D model can be derived from AI Generated images, but it is not ready for most consumers at this point, and I can't vouch for the actual quality, but I am betting some day soon it WILL be a common thing.
Such will certainly work, especially with a blueprint for a shape, like humanoids, or 4-legged animals. Arbitrary shapes... maybe there'll be some experimental/fun stuff, but that's another league by magnitude(s). I'd question how useful it actually is, because i'd assume you'll have more streamlined and fast tools to create 3d-textures for models anyway, e.g. based on a 2d-example.
@Gordig
There are already twitter threads where ”traditional animators” are furious about the Corridoor crew
restyling their live footage to look like anime.