I get this, but I'm kinda new to Photoshop and I have even tried using filters, like you said, before attempting to making that image with line art, I just didn't like the result, so I ended up with what I got. Although, to be fair, I know there must be a perfect combination of filters/settings that will get me to the cartoony style I'm looking, I just couldn't find it yet.
-snip-
Anyways, I don't wanna get hung up on this. So if you know of any Photoshop magical filter combination that can get me from a 2d render to something similar to the above, it would be of great help to speed up this process. By avoiding the hand-drawing grind, I wouldn't mind putting up the extra work into making more renders and even animating with many frames + some frame interpolation AI to create some smooth animated scenes.
Okay, well, if you do want to go for a cartoon style cell-shading? If that is the goal, rather than something you're just settling on? Then you definitely can do that at the render level, and you absolutely should do that. Genuine toon-shading or cell-shading can't be done with a simple Photoshop filter. I know for a fact that more robust and general purpose 3D modeling and rendering programs can do it, and a quick Google search for 'toon shading in Daz3D' brought up a link to their resource store where you can just purchase a working toon-shader for it.
You must be registered to see the links
Now that is built around nVidia's Iray tech, so is gonna require one of their GPU's and probably a not insignificant amount of GPU grunt depending on how you setup a scene. Again, the export resolution and the complexity of the lighting solution would be the major determining factor to your render time here. I kinda doubt you're going to be doing a lot of procedural physics simulations. Using less complex lighting and lesser density mesh models with smaller textures and simplified material properties will also reduce render time.
Also with this pre-packaged option here you are just paying for convenience. If you don't have a nVidia GPU and don't mind watching some tutorials on how to import 3D models and rigs and how to setup some lighting and shaders? You could do all of this with any GPU in any mainstream general purpose 3D model and render application. Plus if you did jump to another render tool, like Blender or such, they tend to be super well documented. Type 'Blender tutorial' into YouTube and you can watch stuff for days without end.
Also, as someone who studied animation at the college level (Game Art & Design, Art Institute of Pittsburgh), let me give you my boilerplate warning about doing animation poorly. Okay, so there is this thing called the
You must be registered to see the links
. Human empathy response scales with something's likeness to being human, the more human it is the more easily we can empathize with it (this is why Wall-E has 2 big camera eyes and they move like eyebrows to help convey emotions). But there is a phenomenon where once you get close too close to being human, without being human enough, and the empathy response nosedives. This effect, when portrayed as a line graph, looks like a sharp valley, with the empathic response rising again as you get closer to a healthy human being. The valley consists of creepy androids, zombies, and human prosthetics. It is also inhabited by anything that attempts photorealism without the animation to match. This has even cropped up in major Hollywood films like
Final Fantasy: The Spirits Within,
The Polar Express, and
Beowulf. That is because animation magnifies the effect, both making less-human beings more empathetic (the aforementioned Wall-E is a good example), but it also can further highlight just how inhuman something actually is (think shambling zombies). Animation is, therefore, a double edged sword.
The more inhuman you are, the more leeway you're gonna have. If you're working with a heavily caricaturized human model like you'd see in Pixar's
The Incredibles, they work much better with a more classical exaggerated 'Disney' style of animation. Even if the animation is poor, you won't get the same creepiness backlash with it as you would a photorealistic model. Now just slapping a toon-shader on a model otherwise intended to be photorealistic? Probably not going to get you much wiggle room. If your animation isn't top notch, it's going to look creepy and weird. No amount of AI interpolation can fix bad keyframes. There is no 'Easy Button' to make good animation. Even motion capture data is a starting point, not the end product. So you really should think long and hard on if you want to commit the time, effort, end energy to doing it right; of if that all would be better spent on just honing your still shot composition instead.
I'm not joking. Look at the VN
Crimson High if you want a crash course on creative use of scene and shot composition. It's basically the only Koikatsu Party based game that I like, and it's because the dev took the time to make pretty good looking and visually distinct models that hold up well at all angles (or at least the one they're willing to show the audience). They really shine in dialogue scenes where you can see what a character thinks or feels by their nonverbal reaction to the ongoing conversation. They often have large groups of characters carrying on a conversation, and the writer never needs to step aside and tell the audience what such-and-such character thought about what was just said, cause you can just see their reaction to it. Instead of using up their development time and energy doing animations, they've instead plotted out an entire shot sequence for a given conversation in order from start to finish, so they they can convey the flow of the conversation visually just as much as they can narratively. It takes a lot more work than just going back and forth with a shot/reverse-shot, or just zooming in on a single character's head like you're trapped in a Bethesda RPG, but it pays major dividends. It takes effort and dedication, but what you don't have to do is teach yourself the fundamentals of animation.
