The DAZ 3d to Character to Movie cheat sheet
SasquatchIsCool
Posts: 262
I've been using AI for quite somet time and have been an early adopter. It is just not getting to be about 80% production ready. If you play around with it you will get great results. I learned that AI and DAZ3d is a match made in heaven if you think outside the box.
You can literally create all types of characters, enhance them with AI and then pose them or animate them anyway you like. I've been using it to make a series about Elves but went back on some of my old Daz3d characters and made a little magic with them. DAZ allows you to customize and create figures that you craft and then you adjust and turn into whatever style character you like. You can then use that character to make movies, mangas, comics and renders.
Here is a DAZ3d cheat sheet. I'm creating a full training coming soon to show you all the POWER of DAZ3d, Midjourney, Unreal Engine, Cascadeur, REal Live actors and Video AI systems.! This is going to be fun.
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
Looks good, but remember, that she is not YOUR character anymore.
She is way different, from what you designed by randomnes and this AI character is a mix of characters taken from uncountable online resources.
"Taken" means more or less stolen, because most of the owners of the source material never gave their intentional consent for use of their property for AI machine learning.
Generative AI is pretty much a deal with the devil.
Sweet! Mixing those tools is fun, I would love to see your training!
This sounds very interesting! AI is like Marmite - people either love it or hate it, or they're intrigued by it but haven't tasted it yet. The only thing I really want to use AI for is inbetweening. I have key images with a start pose and an end pose. It would be amazing to have AI complete the inbetween movement of figure and background to give me the full animation. Have you come across anything that achieves that with a high degree of success? Extra big bonus points if it's offline of course :)
You might be able to use video generation with a start image and end image. then you could pick the individual frames in-between the two images that you want, and use AI to enlarge/enhance each of the video frames you picked. FramePack Studio AI might be the best video generator to use, as it doesn't change things a whole lot from your start and end images, plus it's fast compared to other available AI video generators. Just keep in mind that the AI generative process will not reproduce your original images with 100% accuracy. However, they do typically improve the realism of your start and end images.
If you want to do this on your own computer, for ease of use/installation it's hard to beat Ponokio Browser. It is just a "web" browser with numerous scripts that give you the ability to install many of the open source AI models/generators and run them on you computer. You can install FramePack Studio AI with it, and an image enhancer/up-scaler. There is also an AI image generator plugin for Krita. Krita is an open source Image painting/editing software), the plugin is called Krita AI Diffusion. It works really well for AI Image up-scaling.
I'm sure others will have other options that might work equally well or better. These are the two I'm familiar with.
Well said. It makes things easy, so the ethical implications get locked up in a closet by so many adopters. Sad , really. "Just because you can do a thing doesn't mean you should do that thing."
Thank you. Appreciate all those avenues for investigation.
I wouldn't say that. I follow the current laws and that character is still the same character. I made the character in DAZ3d making it already copyright. Any likeness to her is going against the copyright I have on the actual character.
Almost done with it! Almost ready to upload.
Right now Runway ML lets you literally act in front of the camera so the basic mocap. Midjourney also lets you take your image, upload it and animate it. Sit and act something out onn a video and then replace yourself in that video in Klling AI and it will be in that pose. 3d and AI work really well together.
The thing is everyone's ideas of ethics is different. I could easily say AI learned how to draw Naruto scrapping all the pictures of Naruto. But the vast majority of those pictures were unsanctioned art posted on a website like Arstation. Did they have the rights to draw someone else ip, post it and in many cases use it to PROMOTE and MARKET themselves. Technically it is illegal to do that. It is just the companies have usually turned a blind eye.
Google scrapped data. They won that case.
If Hollywood is going to do it and people will still watch....at the end of the day I'm not better but I do want to see my things come alive. Hence why we even use DAZ in the first place.
Think all those years artists though DAZ3d users were hacks and that all they made was slop. I remember no matter how well composed or lit it was like a photo that you couldn't post it in places like CGtalk. You were ostracized by those groups. Now many in those groups want empathy despite never showing it themselves.
Aside from that I don't think I'm convinced it is stealing.
I'm looking forward to seeing your sheet sheet!!!
Yeah, as human beings, the ability for cognitive dissonance seems to be a very common part of our make-up. We are astonishingly capable of being very malleable when it comes to doing what we want or what can make a profit for us -- we are masters of justifying what we do. History is unfortunately rife with examples of such convolutions going back millenia. AI is just the latest technological advance to make it easier...
that is true for everyone
for E V E R Y O N E !
Simple fact: AI is creating it, not the user.
The user only decides, which "piece of art" to pick from.
It's just like asking a friend, who's an artist do do some things
He comes up with some ideas to pick from
Who deserves the credits?
I never understood the difference between one artist's intensely studying another artist's work in art school, and a digital computer doing the same thing.
The artist who took pen or brush or... to paper deserves the credit, of course.
Do you draw or paint or sculpt? That might help you understand a little better.
Well - unless you can produce art by simply staring at a blank piece of paper - from the humble charcoal stick on up, art is created as an interaction between an artist and their tools. As the tools become more complex they become capable of generating images with extremely little input from the user - and that's true for Daz Studio and for cameras too. But that doesn't mean that an artist can't create their own unique work using those tools.
In the extreme case you can create the bulk of the assets the tool uses - with Daz Studio, you can sculpt your own models and sets from base polygons, paint your own textures etc. With generative AI tools you can train a LoRA based on only your own artwork so that it generates a variety of finished images in your style based on just a sketch you give it. But these tools also come with a bounty of other assets that can make that creative interaction more productive and allow a much wider range of exploration or ideas that was practical before - and then you get the questions of who deserves credit ...
To use your "ask a friend" analogy: if you had 10 artist friends and you gave them a render you created in Daz Studio and asked them to trace it and create an image based on that? Then you mix, match and blend elements from those images and your original render ... who deserves the credits for the final result? You? The artists you used? The artists who created the Daz models and morphs you used for your character and sets? The artist who came up with the iconic movie poster that inspired the pose in your creation? The Iray programmer who determined the pixel values in "your" render based on how you arranged stuff?
If all those other folks individually put way more hours, blood, sweat and tears into the assets you used, than you put into creating your final render (NOT your final render :-)) - what credit do you deserve? But it's still your unique creation.
I've been thinking about doing the same thing and have started making a few head and body shots for upload should I decide to pull the trigger. I'd like to try it with a few stories I've been working on. Maybe even dabble with the video capability as this takes off further.
I agree that it is indeed a deal with a devil. Afterall, the 'you will own nothing and be happy' people are not going anywhere. So might as well have fun until the other shoe drops, as it were. Maybe I'm too fatalistic. I'm reminded of an old Russian joke:
An Englishman shouts "I will die for the crown".
An American shouts "I will die for freedom'.
A Frenchman shouts "I will die for love'.
A Russian says "I will die..."
Cheers!
The artist is inevitably forced to interpret what their eyes see, and no matter how precisely they detail the observation, they never use the original object molecule by molecule, or surface/area by surface/area as input. In contrast, AI captures the object bit by bit, pixel by pixel. The fact that it then delivers recombinations or interpretations of that input is the excuse used to claim it's not someone else's art. But its raw material is the exact—or nearly exact—digital object, not an abstraction refined over years of coordinating vision, limbs, and brain.
I think, you are nailing the basic idea.
AI gets confused with art, as long as someone never shared the experience of investing effort to create something. AI enthusiasts appear to be blind to the artistic process.
I don't think that's really the case - the actual pixel information a generative AI system trains on doesn't really survive any deeper into its neural architecture than it does in ours. The generative AI system is forced to learn an abstraction - the "latent space" - from the input data. Then you give it some descriptive text and/or some rough image (outline, blobby shapes, color palette ...) and a healthy dose of noise and ask it to hallucinate an image in that noise based on that input and what it's "seen" before (like looking for a face in a cloud). It's not an abstraction informed by physical interaction with the world (or tawdry concerns like numbers of fingers :-) (okay, that is getting better)), but it is one that's informed by exposure to more visual data than I could ever hope, or want, to ingest. Yes it can generate "AI slop" till the cows come home if you let it, but as an interactive tool that can help in your own creative process it's pretty interesting.
I was leaving this thread well alone because we already have a thread where I am sharing my own experiences with Ai generated animation
but since I am a moth and there is a flame here
Ai does produce very photorealistic results especially in movie format
something no artist bar a cinematographer with actual human actors can do
itts not what I am doing, I am just animating my DAZ renders with it
but, the stuff I watch on Youtube by others is truly amazing
and since it is mimicking real life trained on movie footage largely
I see a lot less ethical issues with it than churning out art styles
https://youtube.com/@jackvideoai?si=KPGAwgoRyXxZNV4B
is agood example of what I am talking about
Looking forward to seeing it! I've been thinking about trying such a thing out but my extreme lack of free time has meant progress is slooooooow. This sounds like it will a huge jump-start!
What a nice non-response. Do you develop digital systems? That might help you understand a little better.
Respectfully, I offer that if one thinks AI art does not require effort, that there is no investment nor creation, then one is woefully ignorant of the thing one is criticising.
Take Suno, the music generator, for example. Personally, I sweat over the simplest of things, burning through a thousand credits and hours and hours to get specifically the sound or mood that I want. By the end, the song is an expression of precisely the idea that I had in mind. AI only freed me from the constraint of actually having to be able to play, say, a trumpet, or owning one, or having a studio to record in. Virtually everthing important about the song will have been a direct result of my creativity and active volition.
If one can critique that, then the argument is hypocritical coming from someone that uses a digital computer to render art. One critique cannot be valid and the other not.
I believe your critique is only valid for a small percentage of users of AI, the cariacatures, and there is even a name for the class of user you are referring to and the artifacts they produce: "AI Slop".
The others are using AI as a tool to enable their creativity by just removing mechanical barriers, just like DAZ Studio and iRay do. I see no difference at all.
Probably because I've used AI to my advantage more than I chose to just criticize it and risk irrelevance. Look, I'm a software developer by trade and so there is no one here whose job is more at risk than mine. Companies are chomping at the bit to lay off expensive engineers, and it is already happening. But when all the hysteria is over, there will be software engineers because it is not software engineers versus AI, it is software engineers who use AI versus software engineers who do not.
Throughout the history of technology from at least the beginning of the Rennaissance, that has been the case; one ignores that at one's peril. And I note that we all still have jobs. That's because AI is going to destroy jobs, yes, of course, but it is going to create ones that we haven't even dreamed about yet. Much like, before the 1940s, there were no software developers and a "computer" was a person.
I've started a response several times, but I've been unable to find the proper words. Your experiences with Suno seems to mimic my own with image generation.
Getting the image I want requires MUCH more work than typing in one simple prompt, hitting generate and calling it good. I go through an iterative process using a combination of different SDXL checkpoints and Flux to get the style, look, feel, and mood I want for the image. Plus numerous iterations (typically 100's) with prompt tweaking, hand painting of the image, and in painting to finally arrive at what I'm looking for. I often start off with text to image generation, but soon switch to image to image refinement, which almost always involves "fixing" the image to take things out I don't want, or fixing generative errors in the image (things that if not "fixed" will often continue to influence the image in a way I don't want). One thing that really helps the process is after say 25 to 50 images without changing the portion of the prompt that "generates" facial features, the AI will lock onto and more or less keep the face I want (plus using and image to "refine" will keep good consistency of the character). A final image can easily take me 10-40 hours to finish. That may be in part because I'm still learning, but I think that may just be due to the complexity of the process, and I've become more critical of the final outcome.
"AI Slop" - what a great term. It reminds me of NVIATWAS in the DAZ/Poser world. Yes, there are some outstanding NVIATWAS out there, but there were sooo many that simply were poorly done without any artistic thought involved. Now I look back at my early attempts with AI, and I definitely see AI slop! I still haven't been able to get a great image from DreanUp at DeviantArt. Installing the Krita Diffusion Pluging on my own system was a game changer for me. Before that, I couldn't understand how people were creating jaw dropping images. I'm not at the jaw dropping level yet, but being able to just play with different aspects of generation without worry of burning up all of my "credits" really made a difference.
There also seems to be a lot of confusion over how AI images are generated. Saying that AI captures the object pixel by pixel implies a direct capture like a camera, but most generative AI systems don’t do this. They generate images based on learned patterns and statistical probabilities, not literal pixel-for-pixel replicas. Although there are methods for generation digital image replicas of real objects, but that typically involves a real effort on the part of the user to do so (i.e. using LoRa's or specific keywords within a given checkpoint). But the same can be done by very skilled artists too. So AI is not alone in this capacity, and actually kind of a late to the party.
I still enjoy using DS, but there are two things that I prefer with AI. One is pretty simple, I had surgery on my right eye last Jan. and I'm still trying recover. So I'm one eyed right now, working on a 17" laptop screen, and find it quite draining trying to read everything (unscalable fonts) in the DS interface. The second thing is pretty much the opposite of what people often say about AI vs DS. I find DS rather limiting because without creating my own content, I have to compromise what I see in my minds eye vs what is actually available. Sometimes there is actually a character(s), outfit(s), hair, props, and environment that are very close to what I want, but that's rather rare. AI can be difficult to get exactly what you want as well, but I'm not also trying to work with different levels of quality/realism with the items in the image. Plus, I can create things with AI that don't exist in the DAZ world. But the big thing for me is it has allowed me to continue to be creative and learn something new on the computer while my vision is compromised.
Very well said.
I feel the same way about using 3D assets I did not create. I am not blind at all. I am a traditional artist, I have sculpted, sketched, metal work, pastels, ink wash, etc. I also come from a game design background creating 3D models for a couple of game studios and mods for a ton of games. Using DS to pre load ready made assets you purchased from the store and hitting the render button is no different than typing prompts and hitting the generate button in my experience. I put a lot of work into the images I create with AI. If a person chooses not to use AI, then I am fine with it, but I am tired of people bashing those that choose to use it, especially ones that have never tried it.
Thanks for the link Wendy! He (?) has some pretty amazing videos there.
That is I take I agree with. I personally see an opportunity and don't want to be crushed by AI cause it will make parts of my job irrelevant unless I work with it. If we don't we can just go do some manual labor jobs from here on out.