The only thing better than Disney’s magic is their lawyers. Coming from a studio who purchased the Star Wars universe, refuses to pay the authors who wrote hundreds of books before the acquisition but still freely uses those stories, I’m not surprised at all.
yeah but a lot of that stuff has fallen under public domain. Snow White was first published in 1812 so... If public domain laws are life + 70 years or whatever it's fair game
@@shaunmclellan1877it’s actually 95 years, not even the first Mickey Mouse short is in public domain, but the date is closer. The Disney Snow White is transformative work so I guess the same 95 years rule is applied
@@PanConQueso001 We'll see. Disney has already tried to sneak some copyright extensions into major US legislation as pork, I suspect it's only a matter of time before they succeed and copyright becomes some nightmare where the term lasts 125 years.
@@shawnwilcowski Some of that energy itself isn't Disney originated. Many times if something comes up and the patent people get involved you might have to fight someone who really doesn't deserve it, or you risk losing it. As much as I hate Disney, FL resident here, sometimes its actually the government that gums things up. Paris Hilton is an example of this. A company/person was using something she had patented. If she didn't sue the company she would lose the patent. Companies aren't thrilled about projecting an image that screams 'oh look at what an absolute dickhole I am, suing this small child'. Copyright is just fucking broken. That being said, I hope Disney dies in agony and suffering, alone in the dark, with its corpse falling into the deepest abyss known to man.
Minor correction: any motion capture tools still takes manual work to polish and clean artifacts etc. The tools cut thousands of hours, but many videos imply it eliminates it completely.
Agreed! It’s not magical (although closer to it). Plus there’s a whole slew of prep-work: the target final animated model needs to be, well, model but also textured properly, be able to support the range of facial animations, etc that the face capture will drive. Also there’s all of the tracking, merging with the live action actor, compositing, and more. Heck, I bet they go frame by frame and paint on them in many of the shots.
Character animator here; animating high fidelity human faces isn't "tedious", there are plenty of talented artists in the industry who are capable of manually animating an on par job with performance capture; it's just a matter of studios rather spending the money on automating the process instead of paying artists. Benjamin Button was an impressive tech piece for the time, but does look stiff and dated today; while I know individuals, if given the time, could have animated it to hold up better today. That's what character animators at VFX studios do for a living lol, but studios love to cut costs and shorten production schedules any chance they get
Ehh, I'm not buying it. There's some very talented animators out there when it comes to faces, just look at any Pixar film; but examples of this always come from stylized characters. Realistic facial expressions, with all of the incredibly minut details and micro expressions that make a performance hit, are not something I could be easily convinced can be done by hand.
@@godofthecripples1237 just consider the performance capture from every shot in avatar has to be cleaned up by animators, if not re-timed or re-animating some of those performances; some VFX animators are far higher tier then those at Pixar
@@emackenzie the best animators I know tend to love animating lol; you think Michelangelo would have undertaken the Sistine Chapel if he found painting tedious?
There are some things that weren’t here that are important to know. MOVA as it was later known didn’t start with Steve Perlman. Paul Dbevec a USC professor and the research of his grad. students at USC are largely responsible in one way or another for almost all of the camera based tracking for 3D mesh manipulation software that available today. Because it was created at a non-profit University to begin with it’s highly unlikely that that any one company owns the exclusive rights to it. Moreover, at the end of the day it creates a point cloud of animated data points that can be tracked and stored. How the tracked data is interpreted on the other end and applied to the 3D model is likely open to interpretation and any company with the right professionals can do it.
MOVA is one of the rare software that only works in Autodesk Maya, MotionBuilder & 3ds Max. I think this is one of the biggest features that make these products an industry standard and many powerful software used today have not reached peak points for nothing...
Even professional motion capture technology with rigs and cameras requires a lot of manual corrections and fixes. This technology is great and fascinating, but rarely delivers an acceptable result *straight* away.
The real answer is you don't have to use sophisticated camera lens and a tool that costs too much on its own, if you have a budget to get an iphone xr get one, then train yourself how to create and transfer face expressions onto a character for free, there's a lot of tutorials in youtube nowadays for free, instead of photorealism we can go with the opposite direction which is npr(non-photorealism) techniques with a good amount of cel shading, shadows, highlights and etc. NPR will always definitely beat photorealistic ones in the long term and can make more memorable characters. OR you can just make semi-realistic ones.
It's a purpose for small studios but studios like marvel has money and they invest for the highest quality to achieve. But in the next upcoming years these technologies won't look hard other low budget software companies will also improve themselves. As everything in the world depends on money, what you pay what you get
You're talking about 2 different styles. And this is a hobby point of view. Professionally, we seek for solutions to cut costs, save time, tell a story and achieve a professional result. Believing npr beats photorealism is like looking at hobbies comparing two different softwares.
Dimensional Imaging has been developing essentially the same tech as this since 2007. How it is being used has evolved over the years, but even this sort of motion capture process is not without flaws. This tech is best used as a 1:1 recreation of an actors face with motion, the data does not lend itself well to retargeting to different shaped heads (e.g. Hulk), hence why the more tried and tested approach of using FACS face rigs is the most common choice for studios. It is a great technical achievement for sure, but it most certainly doesn't provide shippable results raw. Cleanup is needed and the tracking process is mostly manual. (if you want to map a uniform mesh to the point cloud data that is).
Interestingly, with MOVA, once an Actors face has been captured, realistically that actor will not be needed for many parts of Future Projects. Other face actors could just do the work and drop the emotional spectrum onto the Famous Actor's face Real Fake style. The money that can be saved and the profits that can be achieved are going to be massive.
That's why lots of actors are not allowing their faces to be used on anything but the current project. All other uses have to get license from the actor. As it should be, you cannot use someone's likeness for free.
I dunno; I don't think it's that easy. Basically, your face would have to map perfectly to the actor's, and I don't about you, but my face is not Brad Pitt's...More seriously and detailed: sure, one can say one could readjust the final mesh of one actor to make it "fit" another actor's face, but that's not really the point. It's more that, since our individual facial structures make for our individual expressive natures, what you are gaining through this technology is precisely that: unique facial "contortions" (expressions) as performed by this or that particular actor. That's nearly as unique as a finger print...Now, that being said, what IS a major issue is when a LOT of such recorded expressions by a number of actors (famous and not so famous) are put together into a database and are then ALL used by machine learning to help an "AI" come up with unique "expressive capabilities". This is where things get more interesting, and dangerous for job security in the long run.
'Instantly achieve realistic results with ease'. Anyone who works in the Vfx industry knows this is not the case. The technology is constantly improving, but with the improvements come stronger competition to make higher quality Vfx in the budget - so it's back to square one.
Movie Studios: This tech to transform real life actors to animation realistically is too powerful amd dangerous. AI: Hey world, we don't need real life actors anymore
Indeed...and, by using ALL these recorded "expression performances" into a large database for machine learning algos to learn from...Hence, the big debate about actors and their unique "data likenesses" being used by the major studios without their consent.
See no benefit to this compared to other facial motion capture solutions. The ones currently available to public for free or next to nothing don't require makeup my man :)
It's a multitude of decisions, not just one. The first is Art Direction, the 2nd is a TV schedule/Pipeline, and last will always be budget TV brutally fast compared to movies. I work in TV animation, and my quota to get my shots done for a week, feature animators will get months for. Feature animation the quota on average is 3-7 seconds/week. TV is 35-60 Seconds/week Expecting film feature quality VFX on TV shows is a challenging task, there will always be compromises. But animating a Photorealistic Face, with proper Simulation, Lightning and Rendering, without the proper Time/Budget/Talent, there will always be a dip in quality. Having a realistic face is the Hardest thing in VFX. It's not just a matter of the software the VFX houses use or don't use.
hehehe "The end" anyone with a degree of logicial thinking new it was the end the moment they saw a fluent conversation with a language model,... or from the moment they saw stable diffusion producing somewhat coherent images based on text. I will keep modelling and sculpting and messing with materials and nodes,.. cause tis fun.... in terms of career path though,.. animation is just the next on the chopping block after illustrators,.. and then media in all its varients,.. and then contact center staff,.. teachers,... writers, doctors,.... welcome to the rabbit-hole,... I said it after I saw that first semi coherent image from stable diffusion...... this,.. will make 90% of jobs obsolete
AI will eventually do everything much better. Ironically, all the computer nerds that helped bring her online will be turned into a miasmic blob and used to fuel moon missions.
Great content! Just your speaking style is crying for spontaneity. You are sounding like CGV to me. Please, be natural. People don't speak always with the same prosody repeated on and on the same way. I suggest you take a course on music composition, or take a walk in the forest or by the sea and listen to what godess spontaneity has to say.
You'll have wrinkles whether you want them or not. Wrinkles could be cause by normal maps or even by mesh deformarions. It's a matter of how much you'd like to have depending on your art style. Disney characters for example, would not need as many wrinkles as the Hulk in which they are trying to achieve realism. You want a realistic character without wrinkles? Don't expect to create something that would wow people.
@@Cola-42 People don't need to make hyper realistic stuff on most works unless they wanna make it with live action, or they can just make a very bony face base mesh for the whole face. Or to save time they can just make a physical and a digital bony skin. The bones I'm talking about are the ones that make 3d models move. Also, people don't need to make it hyper realistic to make it stand out, most people in the industry of hollywood nowadays can't even make an awesome story unlike before. Who even needs to make realistic proportions if the theme isn't even realistic in the first place? Making things realistic doesn't change the tone of the story. Also, not everyone's going to point it out when the theme isn't reality even. Remember that spider man multiverse movie? it looks better than the latest disney movies just look at the latest iterations of peter pan and the little mermaid they look so ugly, the whole video looks dead, it feels like its not even made with the intention of making an interesting story. Disney's film industry for animated films looks quite dead.
@@nushia7192 This is a bit of a loaded answer. Making Movies/TV shows is a completely creative process. Just because there is a dip in quality writing from mainstream Hollywood, doesn't mean that creatives should also compromise on achieving the best. It's a very lazy man's way to think.
@nushia7192 Well, I'm only talking about the art style. Not about Disney, Hollywood, or the film industry. If you make cartoon characters, you wouldn't need as many micro details as a realistic character. If you want to specialize in realistic characters, then you need to make them like real-life people, and real-life people have wrinkles. I'm only talking about the people who make characters. Character riggers, animators, and whatnot like you mentioned above obviously have other concerns based on what they do. A lot of movies do have realistic character heads replacing real actors. Most people would not notice until they demonstrate how they made them. Deepfake may replace some of this technology in the future, but right now they still use 3d created heads. The wolverine has scenes where they replace his head, and something like Thanos is obviously 3d made with lots of HQ wrinkles most of our computers can't even handle. You need wrinkles for all kinds of stuff. Do you want to make cartoon dinosaurs or realistic ones? Some people need to be able to make realistic stuff, and the devil is in the details. Your eye is trained to spot those things and tell it's not right.
More AI panick nonsense with a clickbait thumbnail I see. Love how much AI hype has died down now that people have realised that AI tools looks like garbage without a pro to massively clean up and polish it
ActionVFX: BEST Stock Footage For VFX ► bit.ly/3T7UxHv
The only thing better than Disney’s magic is their lawyers. Coming from a studio who purchased the Star Wars universe, refuses to pay the authors who wrote hundreds of books before the acquisition but still freely uses those stories, I’m not surprised at all.
yeah but a lot of that stuff has fallen under public domain. Snow White was first published in 1812 so... If public domain laws are life + 70 years or whatever it's fair game
@@shaunmclellan1877it’s actually 95 years, not even the first Mickey Mouse short is in public domain, but the date is closer. The Disney Snow White is transformative work so I guess the same 95 years rule is applied
@@PanConQueso001 We'll see. Disney has already tried to sneak some copyright extensions into major US legislation as pork, I suspect it's only a matter of time before they succeed and copyright becomes some nightmare where the term lasts 125 years.
Oh not to mention they’re now suing/copyright striking COSPLAYERS!!!
@@shawnwilcowski Some of that energy itself isn't Disney originated. Many times if something comes up and the patent people get involved you might have to fight someone who really doesn't deserve it, or you risk losing it. As much as I hate Disney, FL resident here, sometimes its actually the government that gums things up.
Paris Hilton is an example of this. A company/person was using something she had patented. If she didn't sue the company she would lose the patent. Companies aren't thrilled about projecting an image that screams 'oh look at what an absolute dickhole I am, suing this small child'. Copyright is just fucking broken.
That being said, I hope Disney dies in agony and suffering, alone in the dark, with its corpse falling into the deepest abyss known to man.
Minor correction: any motion capture tools still takes manual work to polish and clean artifacts etc.
The tools cut thousands of hours, but many videos imply it eliminates it completely.
give a scenario, because I have used a similar tool and we rarely touch the facial motion data
@@SkeleTonHammer That's what happens when they let the old men with the money have an opinion and give "feedback" on "their" movie lol
@@SkeleTonHammer Lol what? Go watch those examples again.
@@SkeleTonHammer with that much confidence I can almost believe you
Agreed! It’s not magical (although closer to it).
Plus there’s a whole slew of prep-work: the target final animated model needs to be, well, model but also textured properly, be able to support the range of facial animations, etc that the face capture will drive.
Also there’s all of the tracking, merging with the live action actor, compositing, and more.
Heck, I bet they go frame by frame and paint on them in many of the shots.
Character animator here; animating high fidelity human faces isn't "tedious", there are plenty of talented artists in the industry who are capable of manually animating an on par job with performance capture; it's just a matter of studios rather spending the money on automating the process instead of paying artists. Benjamin Button was an impressive tech piece for the time, but does look stiff and dated today; while I know individuals, if given the time, could have animated it to hold up better today. That's what character animators at VFX studios do for a living lol, but studios love to cut costs and shorten production schedules any chance they get
Why the lol
Ehh, I'm not buying it. There's some very talented animators out there when it comes to faces, just look at any Pixar film; but examples of this always come from stylized characters. Realistic facial expressions, with all of the incredibly minut details and micro expressions that make a performance hit, are not something I could be easily convinced can be done by hand.
People being capable of doing it doesn't mean it isn't tedious work
@@godofthecripples1237 just consider the performance capture from every shot in avatar has to be cleaned up by animators, if not re-timed or re-animating some of those performances; some VFX animators are far higher tier then those at Pixar
@@emackenzie the best animators I know tend to love animating lol; you think Michelangelo would have undertaken the Sistine Chapel if he found painting tedious?
Thanks so much I really appreciate being kept up to date on the ever-changing industry where so many different opinions often jumble together
As a Blender artist, I love this channel.
what you need is a much cheaper solution. like deep fake tech on a phone for most people. that tech looks massively expensive to use.
ask the real thanos if he could pose for deep fake 🤣🤣
There are some things that weren’t here that are important to know. MOVA as it was later known didn’t start with Steve Perlman.
Paul Dbevec a USC professor and the research of his grad. students at USC are largely responsible in one way or another for almost all of the camera based tracking for 3D mesh manipulation software that available today.
Because it was created at a non-profit University to begin with it’s highly unlikely that that any one company owns the exclusive rights to it.
Moreover, at the end of the day it creates a point cloud of animated data points that can be tracked and stored. How the tracked data is interpreted on the other end and applied to the 3D model is likely open to interpretation and any company with the right professionals can do it.
MOVA is one of the rare software that only works in Autodesk Maya, MotionBuilder & 3ds Max. I think this is one of the biggest features that make these products an industry standard and many powerful software used today have not reached peak points for nothing...
this video could have been 5 mins long. Man kept going round and round
Even professional motion capture technology with rigs and cameras requires a lot of manual corrections and fixes. This technology is great and fascinating, but rarely delivers an acceptable result *straight* away.
Right!
Wow , I barely knew about technology used in Hollywood VFX. This is really informative. thank you!!
The real answer is you don't have to use sophisticated camera lens and a tool that costs too much on its own, if you have a budget to get an iphone xr get one, then train yourself how to create and transfer face expressions onto a character for free, there's a lot of tutorials in youtube nowadays for free, instead of photorealism we can go with the opposite direction which is npr(non-photorealism) techniques with a good amount of cel shading, shadows, highlights and etc. NPR will always definitely beat photorealistic ones in the long term and can make more memorable characters. OR you can just make semi-realistic ones.
It's a purpose for small studios but studios like marvel has money and they invest for the highest quality to achieve. But in the next upcoming years these technologies won't look hard other low budget software companies will also improve themselves. As everything in the world depends on money, what you pay what you get
You're talking about 2 different styles. And this is a hobby point of view. Professionally, we seek for solutions to cut costs, save time, tell a story and achieve a professional result. Believing npr beats photorealism is like looking at hobbies comparing two different softwares.
Dimensional Imaging has been developing essentially the same tech as this since 2007. How it is being used has evolved over the years, but even this sort of motion capture process is not without flaws. This tech is best used as a 1:1 recreation of an actors face with motion, the data does not lend itself well to retargeting to different shaped heads (e.g. Hulk), hence why the more tried and tested approach of using FACS face rigs is the most common choice for studios.
It is a great technical achievement for sure, but it most certainly doesn't provide shippable results raw. Cleanup is needed and the tracking process is mostly manual. (if you want to map a uniform mesh to the point cloud data that is).
we need more of these type of videos
Ur videos are always awesome 👏! Thx for educating me dude!
Interestingly, with MOVA, once an Actors face has been captured, realistically that actor will not be needed for many parts of Future Projects. Other face actors could just do the work and drop the emotional spectrum onto the Famous Actor's face Real Fake style. The money that can be saved and the profits that can be achieved are going to be massive.
That's why lots of actors are not allowing their faces to be used on anything but the current project. All other uses have to get license from the actor. As it should be, you cannot use someone's likeness for free.
I dunno; I don't think it's that easy. Basically, your face would have to map perfectly to the actor's, and I don't about you, but my face is not Brad Pitt's...More seriously and detailed: sure, one can say one could readjust the final mesh of one actor to make it "fit" another actor's face, but that's not really the point. It's more that, since our individual facial structures make for our individual expressive natures, what you are gaining through this technology is precisely that: unique facial "contortions" (expressions) as performed by this or that particular actor. That's nearly as unique as a finger print...Now, that being said, what IS a major issue is when a LOT of such recorded expressions by a number of actors (famous and not so famous) are put together into a database and are then ALL used by machine learning to help an "AI" come up with unique "expressive capabilities". This is where things get more interesting, and dangerous for job security in the long run.
'Instantly achieve realistic results with ease'. Anyone who works in the Vfx industry knows this is not the case. The technology is constantly improving, but with the improvements come stronger competition to make higher quality Vfx in the budget - so it's back to square one.
Awesome work. Please make a video on how to we can become a vfx artist using blender. What will be the steps and softwares.
Step 1: Make things with Blender. Step 2: repeat, learn, repeat. Step 3: You're a wizard now.
Movie Studios: This tech to transform real life actors to animation realistically is too powerful amd dangerous.
AI: Hey world, we don't need real life actors anymore
excellent tech journalism. 👌🏽
it's not the tool it's the user
lovely video, I learned a lot through this
The VFX Tool system is countless of people and uncredited Artist doing none stop work correcting this scan data
This is ancient old technology lol
What about Dobby in HP Chamber of secrets which was way before mova?
How about Davy Jones from Pirates Of The Caribbean? Is that using mova too?
Skip to 5:00 for actual reasons and on what's MOVA
so cool, I love the video 🤩
WETA digital already surpassed this.
Unreal Engine new addition will blow this out the fkm water
An awesome video
you should check the new ai face animation trained model from weta, is sick, i read the paper like 6 months ago.
0:30 min is show Thanos head high poly model in Mudbox:::))))))
It’s the end times for artists yep . Even for actors to some degree. Now you can clone voices , faces etc
lol. no
Unable to get the free trial of skill share?
He looks like a Pawn Star.
Thanos was great not because MOVA or he had realistic facial expressions.
Seems like something that will soon be superseded by AI based systems.
Indeed...and, by using ALL these recorded "expression performances" into a large database for machine learning algos to learn from...Hence, the big debate about actors and their unique "data likenesses" being used by the major studios without their consent.
mova was not used for Thanos
6:16 anybody know the films?
dont worry in 50 years there only will be one program used for all LOL
Bro, i like your videos, but sometimes you are extreamly redundant
In what sense?
Yeah he got to the legal issue after 8 minutes of filler in a 10 min video.
Seriously tho
See no benefit to this compared to other facial motion capture solutions. The ones currently available to public for free or next to nothing don't require makeup my man :)
Is that why the faces in She Hulk didn't look as good, were they not able to use that software?
No, it was the art style itself.
@@nushia7192 more so the budget.
It's a multitude of decisions, not just one. The first is Art Direction, the 2nd is a TV schedule/Pipeline, and last will always be budget TV brutally fast compared to movies. I work in TV animation, and my quota to get my shots done for a week, feature animators will get months for. Feature animation the quota on average is 3-7 seconds/week. TV is 35-60 Seconds/week
Expecting film feature quality VFX on TV shows is a challenging task, there will always be compromises. But animating a Photorealistic Face, with proper Simulation, Lightning and Rendering, without the proper Time/Budget/Talent, there will always be a dip in quality. Having a realistic face is the Hardest thing in VFX.
It's not just a matter of the software the VFX houses use or don't use.
Budget and time!! Ever heard of the word Chrunching?
client changed the design over and over, a revolving door of new directors, a deadline that was very ambitious
The tool I created on my channel can do this with ease...just sayin
I'm sad
Unreal engine 5 have this tech in meta humans but have to said it alwais kill my pc
Lost Nuke and now Mova . Lol WTH
it should be leaked to the whole world🤣🤣🤣
hehehe "The end" anyone with a degree of logicial thinking new it was the end the moment they saw a fluent conversation with a language model,... or from the moment they saw stable diffusion producing somewhat coherent images based on text. I will keep modelling and sculpting and messing with materials and nodes,.. cause tis fun.... in terms of career path though,.. animation is just the next on the chopping block after illustrators,.. and then media in all its varients,.. and then contact center staff,.. teachers,... writers, doctors,.... welcome to the rabbit-hole,... I said it after I saw that first semi coherent image from stable diffusion...... this,.. will make 90% of jobs obsolete
it wont lol
AI will solve this soon
If you can't afford it, get bent. Truly unfortunate.
This is an advert buried in ridiculous amount misinformation...
I've never seen a youtube video that just said the same thing over and over again as this video here.
dang you are clueless
Geezus get to the point and stop repeating yourself
in short: "Mova was used" 🤷♂️😅
AI will eventually do everything much better. Ironically, all the computer nerds that helped bring her online will be turned into a miasmic blob and used to fuel moon missions.
Great content! Just your speaking style is crying for spontaneity. You are sounding like CGV to me. Please, be natural. People don't speak always with the same prosody repeated on and on the same way. I suggest you take a course on music composition, or take a walk in the forest or by the sea and listen to what godess spontaneity has to say.
Im so tired of Thanos in vfx tuts.
It's inevitable.
man, why would you make a video on something you know so little about... you know you could've interviewed someone from the industry VERY easily??
Why VFX artists are do obsessed with wrinkles? Seriously, who needs them?🤔
You'll have wrinkles whether you want them or not. Wrinkles could be cause by normal maps or even by mesh deformarions. It's a matter of how much you'd like to have depending on your art style. Disney characters for example, would not need as many wrinkles as the Hulk in which they are trying to achieve realism. You want a realistic character without wrinkles? Don't expect to create something that would wow people.
@@Cola-42 People don't need to make hyper realistic stuff on most works unless they wanna make it with live action, or they can just make a very bony face base mesh for the whole face. Or to save time they can just make a physical and a digital bony skin. The bones I'm talking about are the ones that make 3d models move. Also, people don't need to make it hyper realistic to make it stand out, most people in the industry of hollywood nowadays can't even make an awesome story unlike before. Who even needs to make realistic proportions if the theme isn't even realistic in the first place? Making things realistic doesn't change the tone of the story. Also, not everyone's going to point it out when the theme isn't reality even. Remember that spider man multiverse movie? it looks better than the latest disney movies just look at the latest iterations of peter pan and the little mermaid they look so ugly, the whole video looks dead, it feels like its not even made with the intention of making an interesting story. Disney's film industry for animated films looks quite dead.
@@nushia7192 This is a bit of a loaded answer. Making Movies/TV shows is a completely creative process. Just because there is a dip in quality writing from mainstream Hollywood, doesn't mean that creatives should also compromise on achieving the best.
It's a very lazy man's way to think.
@nushia7192 Well, I'm only talking about the art style. Not about Disney, Hollywood, or the film industry. If you make cartoon characters, you wouldn't need as many micro details as a realistic character. If you want to specialize in realistic characters, then you need to make them like real-life people, and real-life people have wrinkles. I'm only talking about the people who make characters. Character riggers, animators, and whatnot like you mentioned above obviously have other concerns based on what they do. A lot of movies do have realistic character heads replacing real actors. Most people would not notice until they demonstrate how they made them. Deepfake may replace some of this technology in the future, but right now they still use 3d created heads. The wolverine has scenes where they replace his head, and something like Thanos is obviously 3d made with lots of HQ wrinkles most of our computers can't even handle. You need wrinkles for all kinds of stuff. Do you want to make cartoon dinosaurs or realistic ones? Some people need to be able to make realistic stuff, and the devil is in the details. Your eye is trained to spot those things and tell it's not right.
if you take detail out of the face you get she hulk
More AI panick nonsense with a clickbait thumbnail I see. Love how much AI hype has died down now that people have realised that AI tools looks like garbage without a pro to massively clean up and polish it