> Unity’s Demo productions drive advanced use of the Unity real-time 3D platform through autonomous creative projects, led by Creative Director Veselin Efremov.
I was going to say that the overlaps between one lock of hair to another, between hair and background/hair and skin, and the edges of lips and teeth looked a little poorly keyed, like the objects had high resolution textures but the surface map/motion rig was low-poly...which would be annoying but probably easy enough to ignore in an indie film.
But live? On a single (hard to get) consumer GPU? That's seriously impressive. It makes me wonder how much of this is hand-tuned rigging and how much is physics based; if you tried to shake hands with this digital human using a game controller or VR rig, how would that look?
Wait till you see Unreal's Matrix demo running live on a PS5 (even with interactive parts) instead of just some hand-crafted marketing material from Unity on YouTube yet again.
Yes, and did you actually take a look into that? I made the point in another comment: What they provide is either cherry picked assets, doesn't even work properly until you do quite a lot of plumbing or it shows how much is actually "hand crafted" and not using engine specific things. For example you import a captured model and animation that is basically just keyframed vertices, while they claim how cool and procedural their animation toolchain is ... That's not how this works. Unreal does this kind of specific customization for demos, too ... but on a much lower scale. Unity demos barely show actual engine features but are just made to impress the general layperson.
I remember for a long while everything was "Toy Story" quality in real time, the PS2, the PS3, etc. It never really was.
But at some point, we definitely passed it. The room is nifty, but mostly been done. But that is a pretty good person. Lip sync is a bit off somehow... I think perhaps just too overexaggerated in the motions. But I couldn't tell you from a still frame that wasn't a real person, in real clothes.
I also continue to find it amusing that we can build a person like that and sell it as commercial tech, but we still have to record people talking. (Though TTS has taken an interesting turn lately, after years of not much.)
> I also continue to find it amusing that we can build a person like that and sell it as commercial tech, but we still have to record people talking. (Though TTS has taken an interesting turn lately, after years of not much.)
I suspect part of the problem is that dialog in games is still largely "static"; if the writing is pre-canned then it does not make that much sense to try to develop advanced tts to act it out. The situation will become interesting if we manage to produce sufficiently dynamic dialog system where it is not feasible to use pre-recorded voice acting anymore.
Tangentially related: Why can’t movie studios use a SnapChat-like filter to replace the mouth movement of foreign language films with the mouth movement of the voiceover actors? I feel like that technology definitely exists. There are so many great foreign movies & shows but the voiceover + unmatched mouth movement can be so distracting. Is it just too expensive?
I wouldn't want to watch something like that, I want to watch the original film. Then again I also never watch dubbed movies/tv, I prefer reading subtitles.
Yes, motion is highly uncorrelated. E.g. the whole body has no movement when she raises a hand with a chess piece. Humans don't do that. Everything would move from the chest up to assist the motion.
It's motion captured, that's probably just how the actress moved. Real stuff looks fake all the time. The parts that look wrong in the animations are mostly due to small imprecisions, especially in the facial ones.
It occurred to me yesterday that at some point someone will inevitably advocate for the banning of deep fake pornography in their country by creating a porno starring the politicians of their lower house. I reckon it’d be illegal within the week.
Interesting, but I wonder how this would be enforced. I wouldn't be surprised if over half these cases were people who live in entirely different countries. would another country extradite a citizen over this? This has "technically" been possible for centuries so I wonder if there's any case study surrounding this.
It is clear that the Unity devs spent much time modeling the bones, muscles, blood vessels, fat, and unique skin properties of the face. I would not be surprised if surgeons were consulted. Was this attention to detail given to other parts of the human body?
A lot of people are working on it. They mainly live on Patreon. For VR, specifically. It’s the only feasible way to approach 6dof porn, currently, AFAIK.
I mean you could have hired a hyperrealist oil painter to paint porn characters 30 years ago but it is so much more work and expense compared to people.
I don't see how a game engine is much different for the foreseeable future. It is so far away from under cutting humans in this domain in terms of time and expense.
Electric cars/bicycles/scooters/motorcycles, computers in everyones pocket, huge cheap TV-s, fast wireless internet, robotic lawn mowers, radar cruise control and the list goes on and on and on...
depends what you mean by convenience. The image quality of a 4k camera isn't that much better, but that's largely because it doesn't need to be. Cameras in the 80s already produced great pictures. The difference is that they can weigh less than a pound and be run for hours at a time without worrying about paying ridiculous amounts for film.
Hmmm, a 4K camera that can fly around for 30+ minutes with a controller attached to your cell phone (which in and of itself would blow the mind of any person in the 80s) that has the visual clarity of an image shot on 35mm motion picture camera weighing in at 70+lbs that could only be used aerially by renting a professionally piloted plane/helicopter at more $$ per hour rental than the drone you fly on your own costs?
No, I can't imagine any benifits for modern tech from the 80s ;-)
well, high quality digital is a whole different game from film. in the 1980's the closest thing was a VHS camcorder and my 4k micro four thirds digital camera absolutely blows a 1980's VHS camcorder out of the water.
Yes the image quality of a 4K camera is technically less than that of a film video camera, but show a 1980's movie producer a digital workflow and their mind would be blown. You shoot the picture and can then immediately review and even edit.
So 4K video beats home VHS on quality by leaps and bounds, and for anyone who would have been producing video on film the workflow for digital is amazing.
Drones, for sure. Only a few years ago, a decent digital camera cost $400. Now I can buy one just as good, for the same price, that fucking flies.
Anyone who doesn't experience a genuine "Holy shit, I'm actually living in the future" moment when they first encounter a Mavic Mini or similar modern consumer drone is probably in need of pharmaceutical help or talk therapy.
My parents weren't just released from prison after having been locked away from society since the 80s. Nor have they just awoken from 35 year comas. Sure, they'll have a different perspective than someone who doesn't remember the 80s but they'll surely also have a different perspective from someone with a more abrupt and recent introduction to the latest modernity.
They'll have watched The Jetsons, Blade Runner, 2001, 2010: The Year We Make Contact, and a host of other science fiction that all made the year 2000 look like the year 2100. In 1981, we sure didn't think the Berlin Wall would fall and the USSR would cease to exist a decade later (unless that came about by an apocalypse).
My mind was blown out the back of my head the first time I saw Super Mario Bros in an arcade in 1986-7. I wonder what would have happened to my sanity if someone had shown me a modern game one minute later.
I mean, yeah, if they're satisfied with viewing demos on Youtube. The vast majority of people can't actually play the newest games at their best settings :D
The way people use computers and the Internet has been a huge change. Wikipedia sounds like something that couldn't possibly work, except it turns out that it (mostly) does. Self-landing rockets are pretty impressive. I think the rise of free and open source software would be surprising to most people. The fact that Russia and the United States haven't directly fought a war with each other in all this time and our cities haven't been reduced to rubble by nuclear weapons would seem pretty remarkable to an 80's person. (Though one might want to hold off a few weeks/months before declaring premature victory on that front.) Dystopian predictions about the environment were kind of right -- the effects are there, but American cities don't look like Blade Runner quite yet. Manipulation of society doesn't look like 1984 unless you're in an authoritarian country. Instead, big brother watches you from electronic devices that people voluntarily buy and use, and "big brother" is usually a private adtech company.
Your post is being downvoted because blocking cookies does not solve the problem. It is blocking cookies that causes the issue. When cookies are blocked (in my case, third-party cookies are blocked), an overlay appears over the video explaining the website that hosts the video will not allow the video to be played without a targeting cookie. FWIW, removing the overlay was not enough to get the video to play on the website; but I haven't looked any further into it—easier to just go to YouTube to play the video.
This is really impressive assuming it is as automated/scalable as Unity implies.
The skin and hair look great which are both really challenging for their own reasons, but one particular detail that was surprising was at 1:36. If you watch her lips closely you can see where the top and bottom lip stick together as they open. If they can simulate that level of skin detail when the motion is backed by mocap data this could be a huge quality jump for character-driven and dialogue heavy games.
There are definitely still moments where the lips, teeth, and tongue look "3D" but I don't think I would notice if I weren't hyper focusing on the mouth and just enjoying the story.
> they tend to show off how tech will look 4-5 years down the line
Unity's tech demo "Adam" is now more than 5 years old and I still don't see any Unity games coming even remotely close to that. Unity's tech demos are just hoaxes: Carefully crafted demos with custom engine parts and basically hand-written stuff like shaders and controllers all over. The irony is that these demos are primarily aimed at developers, yet you can't take a single piece of those things and just make it work on your own setup. In contrast when it comes to Unreal it's usually just a Ctrl+C and Ctrl+V away from success ...
Unity games? No, not to my knowledge. But the industry has certainly gotten extremely close to realizing the tech being leveraged in Adam. Some may even argue that some UE5 showcases with real samples have succeeded it, despite this being a non-real time showcase.
I don't think this is because you can't just copy-paste an Adam asset into the Unity engine, however. AAA devs are going to leverage their own assets regardless.
I think that's more of an image problem than a "technologically possible" problem, however. Many of Unity's huge successes come from the indie side, the Ori's, the Hollow Knights, the Cupheads. And it has a very big grasp on the mobile market. Even Pokemon Go decided to use Unity despite the IP being backed by a publisher known for their in-house engines. so a AAA studio isn't thinking of Unity for their next Call of Duty, but maybe for their Call of Duty mobile title.
They seem to be trying to combat that sentiment with the likes of DOTS,HDRP, and whatever Ziva is trying to do, but those are still TBD.
rendered hair gets better all the time, but it's still not there.
sucks for me because I am weird and I find it hard not to pay close attention to hair in real life, sometimes. this makes flaws in rendered hair extremely obvious to me. :(
Definitely, the hair effects are still not there. Something is off, the hair moves as a block in an unnatural way. The rest of the face is pretty good though. The eyes are weirdly intense, but I suppose it was a deliberate choice and some people with blue eyes look like that.
I agree about the hair. It looked great when it didn't move. I thought the eyes were great; to the extent that they were intense, she's an intense looking lady; and they may have done so to amplify the effects.
I also thought the lips were great. It was the interplay between lips and teeth that really threw me off.
The mouth movement[0] sold it for me. Mouths are incredibly hard to do in CG, because you have to have realistic interactions between lips, teeth, and tongue. Not sure how they scanned this so well (maybe it was hand animated?) but it looks excellent.
Really? Because that was the worst part for me. To me it looks absolutely and positively synthetic. I mean, it is better than Terrance and Phillip from South Park, but perhaps because it's "close to reality, but not there yet" it falls into the uncanny valley to me.
Perhaps this is similar to how some people aren't bothered by bad kerning (to which I'm fairly tolerant).
You're right. The mouth movement was awful. I have no idea what OP is on about saying it was good. Maybe my uncanny valley is also steep, but nothing about this demo struck me as good.
The tech maybe is good, but I wouldn't know it from this video. The animation and lighting are just awful.
Link in the OP states screen space global illumination was used, which is probably why the part where the character descends below the surface seems really fake. Still very behind Unreal Engine unfortunately, which already supports ray traced "real" global illumination.
"Look at our demo of a hyper-realistic human character!"
eighteen million things happen around the character to distract the viewer's attention from the human character throughout most of the video
Not the most confident tech demo. Looks like a reasonable degree of evolution, but as they start to get up toward turn-of-the-century movie CGI in games, now you run into the same issues that you see in mocap for movies--including that a lot of details of an actor's expressions and movements actually aren't captured, so you have to have animators go back and laboriously add all those lost details back in to have something that looks convincing.
(Mind, I say turn-of-the-century, but man, the fundamental techniques for rendering skin convincingly have come a long way, even in current game engines.)
Someone who knows better can correct me, but I assume they put the character in a complex environment to show that they are rendering a highly realistic character while still rendering a complex environment. I remember very impressive character rending tech demos from 10-20 years ago, but it was a single character in a static environment.
They're showing off a lot of things at once here. The real time raytraced lighting on their vfx is impressive. To do that in tandem with their human while maintaining 30fps is pretty something.
Every thing draws from the same well so to speak, a human that looked a little better but had to exist alone in a white room would have far less utility for real things.
eighteen million things happen around the character to distract the viewer's attention from the human character throughout most of the video
Yes. Why did they do all those crappy CG effects while showcasing photorealistic characters? Two people seated across a chessboard in a realistic room would have been more effective.
It's getting to the point that everybody is doing this. Unreal Engine has Metahuman Creator.[1] Even Second Life has reasonably good heads now.[2] And has facial tracking on their roadmap.
I guess the simple reason is because then you wouldn't believe it was computer generated but it could just be a movie of two people playing chess. It needs to add a touch of the bizarre or fantastic to remind the viewer they are watching something that isn't real. Though, frankly, the talking portion was reminder enough of that.
I didn't think I would care about face tracking, but Star Citizen uses a Face over IP system and man it really enrichens the online experience. You can communicate a lot more subtlety with it.
It also has the benefit of necessarily head tracking as well, so you can also do TrackIR style head tracking which again adds a lot to the experience to be able to move your view around independant of the mouse and in ways a mouse can't move.
Also they have a character making weird facial expressions in a weird situation - I legitimately cannot tell whether the occasional uncanny valley effect I felt was due to intentional direction or just the limitation of the tech. I suspect this is intentional.
It isn't perfect yet, and there are subtle queues that this isn't a real person. The character and dialog do seem to be deliberately leaning into this so these flaws don't detract. To me, the character started moving her hands in a very unreal way at the start. And then I realized the movement I was seeing was almost certainly motion captured and likely exactly as the actor performed it!
What's interesting is that for most of my childhood (80s and 90s), aside from a few wow moments I was pretty underwhelmed by the standard of the tech. It was like it was trying hard to be something it wasn't yet ready to be. I used to walk into television shops and think they all just look crap. Computers used to frustrate me so much - crunch crunch to do anything, and 256 colours was deemed good (!?). The first music players where you might be able to get one whole album onto a memory card that was too expensive to put a price in the rs catalogue (or perhaps too volatile a price). Anyway, tech was crap. Then around 2005 or something, it started becoming what it was meant to be. You could buy a computer and it could do everything you needed it to do; of course you always wanted more number crunching, but you could see where it was only just right the corner. Then GPUs started doing computation, and computation stopped being thought about. Memory was super fast and copious. One now felt limited by programming capability, not hardware. I'm now genuinely excited by technology. As much as it pains me to say it, the television departments are places of wonder.
It's in that context that this post feels like another step towards achieving some promised vision. If this is realtime, it is truly fantastic.
Now to hope we can deal with climate change and despots and poverty so it's not all for nought.
As an aside, one early wow moment was the first time I saw a mini camcorder, then another for the first genuinely mobile phone; both Sony I think. Also, though a bit later, I remember seeing an in-car GPS and deciding it was basically a perfect interface for it's task.
If you're interested in tech becoming really magic, lookup the VRchat club scene. Real-life skilled DJs put out of clubs by covid, setup a twitch stream with a webcam on their DJ controller, while wearing full-body VR themselves, also streaming their DJ software and in-game view to an in-world stage screen in front of up to 80 guests, many if not most also wearing full-body VR, dancing along together, in any kind of digital world and wearing any kind of semi-humanoid (or not) 3D model you can imagine. The "metaverse" has already been running for a few years, but it's not Meta's, it's a thin multiplayer VR wrapper over Unity.
It’s that you’re alone in a room with a screen strapped to your face looking at pixels instead of seeing, touching, feeling, smelling, hearing, sweating, dancing, and sensing in the middle of a rave of physical human beings
It’s literally senseless in comparison, lossy digitalization of basic analog humanity ie dystopian
Yes even voice chat feels dystopian if thats all your interaction with other humans. Social platforms have ironically made us less social beings. Real ourselves live in the digital world while our fake zombie selves do the mundane tasks in the real world. It does not feel right at all. Its easier to connect and find similar people through internet so i get why it is like this, but i think its also the factor thats making us less mentally balanced, causing real world to feel more foreign and harder to interact with. Drug of social interaction that's not real which leaves you lonely in the end.
I would say the cause is that current internet social connection techniques don't have enough depth to them. Telephone doesn't let you see hand or facial emotions of the other person. Twitter/facebook is only using a keyboard + reading text. Instagram you can see a snapshot of an instant and peoples thoughts about that instant. etc...
VR is working to try and bridge all of these by essentially creating a transporter/teleporter to a shared physical space. Imagine if VRChat was a 1-to-1 replication of you. I feel like this is one of the end goals of VR.
Its easier to connect and find similar people through internet so i get why it is like this, but i think its also the factor thats making us less mentally balanced
This isn't true of just internet, this is true of cities. People no longer care about one another, each person/bond is replaceable with another one. i.e dating apps only help prove this fact. We no longer need to rely on one another to survive and therefore are more independent. We are less willing to give up (sacrifice) parts of our independence and less tolerant of flaws in others. I think this is what leads to higher divorce rates now. How many people live in high rise buildings and actually know many of their neighbors?
This is indeed rendered in realtime, but one thing to note is it's a "4D" capture, more-or-less meaning each frame of the animation is its own asset. This makes it possible to reproduce subtle physics like the lips sticking together slightly when the actor opens her mouth. The amount of storage space, alone, makes this impractical for anything other than demos. Unity claims they will be able to achieve this level of fidelity using a deep learning-based compression that will allow stuff like this to appear in game cutscenes, but all the movements will still be pre-baked. The only interaction possible will be moving the camera. At that point the technology will be very useful, but it's still a ways away from having such a realistic character that can react to you dynamically.
(Though whether that's just a couple years of software technology progress, or a decade+ for hardware progress, who can say?)
Unity recently acquired Ziva, which specializes in the detailed animation of humans and other animals. They were known for their (not realtime) physics-based solutions, but now they have an ML model for faces, apparently. As far as I know, it's still in beta and not widely available. Unity says they will re-release this demo with the Ziva face in a matter of weeks and the quality will be even higher. And possibly allowing interactivity as well?? I guess we'll see in a few weeks.
Superresolution. You have a lower resolution animation (less pixels = less calculations) and then use superresolution to turn that into a 4K image. This is reality right now for NVIDIA GPUs ( I think it’s called DMSS)
Nvidia DLSS is an important part of how they achieved 30Hz at 4k resolution, but that's more of a shading assist and doesn't affect the animation. The facial animation will be compressed with Ziva's ML solution.
There is one out there from 5 years ago or so that is similar to Google's Seurat but for animated stuff, I think pre-baking triangle culling for different views within a limited volume. I can't remember the name of it, from the details I remember (there was a realistic orangutan or something like that rendered with fur) I should be able to find it on Google, but Google search has become degraded recently.
Movement won't be pre-baked, a physics engine sim will be baked in to the neural network, and movements will be another dimension for the deep learning network. And then all of that will be baked into an agent that has been trained to carry out motives (with a simulation of your character, etc). The same applies to speech as movement. And the deep-learned compression rate will be magnificent.
No predictions, just an explainer of how AI agents are trained. For instance, RL is about presenting an environment via rules (gravity, etc), and letting the agent learn its way around, thus discovering what it can and cannot do (a policy for the environment).
You didn't explain how anything actually works, you gave a very crude prediction with a lot of holes of how you think something will work in the future.
Cause they want to push the limits and make their engine look amazing. Also, if they research hard enough, in-game becomes nearly as good as video to the point you can’t tell.
Cutscenes work a lot better (more immersive) if they can correctly reflect runtime-defined assets, e.g. your own character with your customizations, gear and clothes, etc, or the dynamic state of the environment in which gameplay was happening: destruction debris, current time of day, and such.
Just google Unity tutorial. If I remember correctly Unity uses C#.
You are not going to be doing anything close to this out of the gate but I do recall it being a pretty easy environment to learn for novices.
318 comments
[ 4.6 ms ] story [ 276 ms ] threadDoes this mean this was rendered in real-time?
by rob cupisz (Tech and Rendering Lead at Unity Demo Team) https://twitter.com/robcupisz/status/1505875759243612172
But live? On a single (hard to get) consumer GPU? That's seriously impressive. It makes me wonder how much of this is hand-tuned rigging and how much is physics based; if you tried to shake hands with this digital human using a game controller or VR rig, how would that look?
But at some point, we definitely passed it. The room is nifty, but mostly been done. But that is a pretty good person. Lip sync is a bit off somehow... I think perhaps just too overexaggerated in the motions. But I couldn't tell you from a still frame that wasn't a real person, in real clothes.
I also continue to find it amusing that we can build a person like that and sell it as commercial tech, but we still have to record people talking. (Though TTS has taken an interesting turn lately, after years of not much.)
I suspect part of the problem is that dialog in games is still largely "static"; if the writing is pre-canned then it does not make that much sense to try to develop advanced tts to act it out. The situation will become interesting if we manage to produce sufficiently dynamic dialog system where it is not feasible to use pre-recorded voice acting anymore.
It will probably become as routine as automatic dialog replacement.
https://en.wikipedia.org/wiki/Hyperrealism_(visual_arts)
Title claims something the blog doesn't, and creates overexpectation
Even those parts?
The Italian senate approves this comment.
I don't see how a game engine is much different for the foreseeable future. It is so far away from under cutting humans in this domain in terms of time and expense.
No, I can't imagine any benifits for modern tech from the 80s ;-)
Yes the image quality of a 4K camera is technically less than that of a film video camera, but show a 1980's movie producer a digital workflow and their mind would be blown. You shoot the picture and can then immediately review and even edit.
So 4K video beats home VHS on quality by leaps and bounds, and for anyone who would have been producing video on film the workflow for digital is amazing.
Anyone who doesn't experience a genuine "Holy shit, I'm actually living in the future" moment when they first encounter a Mavic Mini or similar modern consumer drone is probably in need of pharmaceutical help or talk therapy.
…
“It works sometimes, I swear.”
You don't exactly need time travel to get an opinion from someone who lived the 1980s... try asking your parents?
They'll have watched The Jetsons, Blade Runner, 2001, 2010: The Year We Make Contact, and a host of other science fiction that all made the year 2000 look like the year 2100. In 1981, we sure didn't think the Berlin Wall would fall and the USSR would cease to exist a decade later (unless that came about by an apocalypse).
I am way more disappointed on our societal progress than I am on our technological progress.
But to be completely honest, I was expecting at least a moonbase by now.
Instead, they got Tang
If you prefer not to accept cookies.
An effing stupid move to force cookies on a promotional site like this.
Foot, shot.
The skin and hair look great which are both really challenging for their own reasons, but one particular detail that was surprising was at 1:36. If you watch her lips closely you can see where the top and bottom lip stick together as they open. If they can simulate that level of skin detail when the motion is backed by mocap data this could be a huge quality jump for character-driven and dialogue heavy games.
There are definitely still moments where the lips, teeth, and tongue look "3D" but I don't think I would notice if I weren't hyper focusing on the mouth and just enjoying the story.
I put my ganer hat on looked at their old demos. Demos always look great. I'll believe it when I see it in gameplay footage.
IME they tend to show off how tech will look 4-5 years down the line, about the time a modern AAA game needs to from production to release.
Unity's tech demo "Adam" is now more than 5 years old and I still don't see any Unity games coming even remotely close to that. Unity's tech demos are just hoaxes: Carefully crafted demos with custom engine parts and basically hand-written stuff like shaders and controllers all over. The irony is that these demos are primarily aimed at developers, yet you can't take a single piece of those things and just make it work on your own setup. In contrast when it comes to Unreal it's usually just a Ctrl+C and Ctrl+V away from success ...
I don't think this is because you can't just copy-paste an Adam asset into the Unity engine, however. AAA devs are going to leverage their own assets regardless.
I think that's more of an image problem than a "technologically possible" problem, however. Many of Unity's huge successes come from the indie side, the Ori's, the Hollow Knights, the Cupheads. And it has a very big grasp on the mobile market. Even Pokemon Go decided to use Unity despite the IP being backed by a publisher known for their in-house engines. so a AAA studio isn't thinking of Unity for their next Call of Duty, but maybe for their Call of Duty mobile title.
They seem to be trying to combat that sentiment with the likes of DOTS,HDRP, and whatever Ziva is trying to do, but those are still TBD.
https://80.lv/articles/enemies-the-working-process-behind-un...
sucks for me because I am weird and I find it hard not to pay close attention to hair in real life, sometimes. this makes flaws in rendered hair extremely obvious to me. :(
I also thought the lips were great. It was the interplay between lips and teeth that really threw me off.
That's a sheer compute power problem. If you have a big enough render farm, you can run physics on every strand of hair.
0. https://youtu.be/eXYUNrgqWUU?t=87
Perhaps this is similar to how some people aren't bothered by bad kerning (to which I'm fairly tolerant).
The tech maybe is good, but I wouldn't know it from this video. The animation and lighting are just awful.
eighteen million things happen around the character to distract the viewer's attention from the human character throughout most of the video
Not the most confident tech demo. Looks like a reasonable degree of evolution, but as they start to get up toward turn-of-the-century movie CGI in games, now you run into the same issues that you see in mocap for movies--including that a lot of details of an actor's expressions and movements actually aren't captured, so you have to have animators go back and laboriously add all those lost details back in to have something that looks convincing.
(Mind, I say turn-of-the-century, but man, the fundamental techniques for rendering skin convincingly have come a long way, even in current game engines.)
Every thing draws from the same well so to speak, a human that looked a little better but had to exist alone in a white room would have far less utility for real things.
Yes. Why did they do all those crappy CG effects while showcasing photorealistic characters? Two people seated across a chessboard in a realistic room would have been more effective.
It's getting to the point that everybody is doing this. Unreal Engine has Metahuman Creator.[1] Even Second Life has reasonably good heads now.[2] And has facial tracking on their roadmap.
[2] https://www.unrealengine.com/en-US/metahuman-creator
[2] https://www.youtube.com/watch?v=OEjvD8C0g8w
It also has the benefit of necessarily head tracking as well, so you can also do TrackIR style head tracking which again adds a lot to the experience to be able to move your view around independant of the mouse and in ways a mouse can't move.
Interesting to see that hair apparently is still a real problem. It looks and moves very unnatural.
It's in that context that this post feels like another step towards achieving some promised vision. If this is realtime, it is truly fantastic.
Now to hope we can deal with climate change and despots and poverty so it's not all for nought.
As an aside, one early wow moment was the first time I saw a mini camcorder, then another for the first genuinely mobile phone; both Sony I think. Also, though a bit later, I remember seeing an in-car GPS and deciding it was basically a perfect interface for it's task.
Is Discord dystopian because people alone in their rooms have a headphone fitted to their head "just to be 'together with other people'"?
Is it that it's more immersive than just audio and therefore...somehow bad?
It’s literally senseless in comparison, lossy digitalization of basic analog humanity ie dystopian
VR is working to try and bridge all of these by essentially creating a transporter/teleporter to a shared physical space. Imagine if VRChat was a 1-to-1 replication of you. I feel like this is one of the end goals of VR.
This isn't true of just internet, this is true of cities. People no longer care about one another, each person/bond is replaceable with another one. i.e dating apps only help prove this fact. We no longer need to rely on one another to survive and therefore are more independent. We are less willing to give up (sacrifice) parts of our independence and less tolerant of flaws in others. I think this is what leads to higher divorce rates now. How many people live in high rise buildings and actually know many of their neighbors?(Though whether that's just a couple years of software technology progress, or a decade+ for hardware progress, who can say?)
what does that mean? To me, they might have just as well said middle out compression.
There is one out there from 5 years ago or so that is similar to Google's Seurat but for animated stuff, I think pre-baking triangle culling for different views within a limited volume. I can't remember the name of it, from the details I remember (there was a realistic orangutan or something like that rendered with fur) I should be able to find it on Google, but Google search has become degraded recently.
> Tension tech for blood flow simulation and wrinkle maps, eliminating the need for a facial rig for fine details
Which sounds like it is not 100% prebaked animation?
chuckles
One baby step at a time.
With this, you just have the character do exactly what you want, when you want, without needing to talk to an agent.
Of course most games don't care about 5 years later, but it still looks crap.
Granted, the gliding across an unwalkable moving floor gear contraption didn't help avoid the perception that I was looking at an automaton.