Slashdot Log In
Leaping the Uncanny Valley
Posted by
timothy
on Tue Aug 19, 2008 11:19 AM
from the compare-this-to-the-final-fantasty-movie dept.
from the compare-this-to-the-final-fantasty-movie dept.
reachums submits this glance at "the newest level of computer animation," intended to get past the paradoxical "uncanny valley" — that is, the way animated humans actually can appear jarring as the animation gets hyper-realistic. "This short video gives us a glimpse of what we can hope to see in the future of computer games and movies. Emily is not a real actress, but she looks like a real person, something we haven't truly seen before in computer animation."
Related Stories
[+]
Hardware: Scientists Achieve Mental Body-Swapping 297 comments
SpaceAdmiral notes the news that scientists have succeeded in convincing experiment subjects that a mannequin's body is their own, and even feeling at home in the body of someone of the opposite sex. The effect could prove useful in virtual reality applications and in robot technology. Here's the paper on PLoS ONE.
This discussion has been archived.
No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
We've heard this before (Score:5, Insightful)
There was much talk about the uncanny value when Final Fantasy: The Spirits Within [amazon.com] came out after Square had promised for years that it would have realistic humans. A common criticism was that the human beings were real enough to inspire comfort for long enough that one would be then shaken by their lack of certain flexibility and the bloodlessness of their faces. Dr Aki was more creepy than sexy.
Re: (Score:3, Insightful)
Re:We've heard this before (Score:5, Insightful)
Parent
Re:We've heard this before (Score:5, Funny)
Parent
Re:We've heard this before (Score:5, Insightful)
You're just imagining things. The real Emily undoubtedly sometimes opens her mouth a little too wide...probably moves centered around her nose.
If you pay attention to the video, you'll notice that they flat-out said that it isn't computer generated imagery. They're just doing face tracking without using markers and mapping that to a mesh.
Big deal. People do that now and get extremely realistic (perfect) results. All you need is two cameras.
It would be interesting if they were actually generating models from this that could work independently (i.e., get CGI-Emily to move in a way that real Emily hasn't been recorded as moving), but they aren't claiming that they can do that...so they probably aren't.
Would it be easy to film somebody with a 3d camera setup and shove them into a video game? Well, I suppose there are a few technical challenges in the sense that there aren't any 3d-movie inside-videogame codecs, but its nowhere near as impressive as the headline makes it out to be.
Parent
Re:We've heard this before (Score:5, Funny)
Dr Aki was more creepy than sexy.
"More creepy than sexy" -- four words that sum up most of the anime I've been subjected to. But if that's going to be the criteria under which we judge the depth of the uncanny valley, some people are going to navigate it a lot faster than others.
Parent
Re:We've heard this before (Score:5, Funny)
Yes, but you are talking about anime - the same animation style that introduced the world to tentacle porn.
Parent
Re:We've heard this before (Score:5, Funny)
"More creepy than sexy" -- four words that sum up most of the women I've been subjected to.
Actually, it's probably: "More creepy than sexy" -- four words that sum up most of the reactions of the women who've been subjected to me.
Parent
Yes, but does it even exist? (Score:5, Interesting)
Well, a better question is if the uncanny valley really exists. Or rather, if it's really as simple as that valley, or we're actually looking at a more complex and multi-dimensional phenomenon.
And I'll attempt to build a framework to falsify it. It's a bit roundabout and I'll start by explaining the what and why of that framework, before all else. Bear with me, please.
First of all, before someone jumps in with the ever popular, "OMG, you're not worthy to question the high priests!" (err... "scientists"), the uncanny valley is just a hypothesis. A very compelling and well argued one, no doubt, but hardly a proven fact.
Second, before I get into the meat of the argument, the points chosen to represent it are highly debatable. E.g., is a zombie scary because of being close enough to the real thing to fall in the "uncanny valley", or because of the whole cultural meaning of death, undeath, corpses, etc?
When you look at each point individually, you can handwave and argue it to be wherever you want it, to support your hypothesis. It's called the Texas sharpshooter fallacy [wikipedia.org], after the fable of the sharpshooter who shot first and then painted a bullseye around the hole. You can "prove" anything in (pseudo-)science if you can do just that to the data: take a fuzzy and ill defined points and argue where they belong on your curve.
The "uncanny valley" paper does just that. We don't know the exact X coordinate on that graph for a zombie or a robot. It could be way right or way left, or whatever. So what really follows is that Mori decided a priori where they belong on that curve, and then places them at a point based on that. It's a textbook application of the Texas sharpshooter fallacy.
So what I'm going to do is an ad absurdum reduction of his curve.
I don't know the exact coordinates of any of my examples either, but, here's the important part: I don't need to pretend to. I'll just peg them between two other values, which, assuming the curve is correct, both fall in the valley or outside it, or some other position. Based on the reaction they caused, and, again, assuming that the curve were correct.
And due to the shape of the curve, if two points are in the valley, then everything between them is in the valley too. If two points are, say, both to the left of the valley, then a point between them should be on the left of the valley too. That is the important part.
So, let's build a counter-example: the FF movie was called a clear example of the Uncanny Valley. It's in the valley. Sony's Everquest 2 (particularly with the unnatural ambient bloom enabled) caused a similar reaction, and many euphemisms were used to describe just that: that that world looked disturbingly unnatural, especially if you pushed the graphics settings high enough. Classic example of entering the uncanny valley from the left, eh? So it's point 2 in that valley.
A point between them should, obviously, also be in the valley. That curve only has one dip, right?
Well, point #3 could be Oblivion. The graphics are better and more detailed than Sony's graphics in EQ, but don't even come close to the insane polygon counts and animations of the FF movies. It's between the two points. It should also be in the valley. It isn't. Nobody was repulsed by Oblivion's graphics. Or pick Crysis, or whatever newer high-end game, and you get the same curious behaviour. It ought to be in the valley, but it isn't.
Let's build another counter-example: so we're told that zombies are only repulsive because they're so close to humans as to fall in the uncanny valley. So logically, if you start with a zombie and move farther and farther away from human-like with it, eventually it exits the valley. Right? In fact, past a point it becomes outright _cute_ and appealing. Or ought to. I mean, that's the shape of that curve.
You probably realize already how absurd that statement is, but let's actually imagine it. Let's say we start with that corpse an
Parent
Re:How true was this? (Score:5, Insightful)
Parent
Uncanny in the other direction (Score:4, Interesting)
Parent
Re:Uncanny in the other direction (Score:5, Interesting)
There actually have been theory's put forward that there will be a second 'uncanny valley' when it comes to transhumanism.
In other words, as 'normals' begin modifying their appearance outside of what is considered normal, they will start to slide down that valley and become socially unacceptable till they get to the point where they are back on the other side of the valley and just accepted as 'freaks'.
Parent
Re:Uncanny in the other direction (Score:4, Insightful)
Parent
Re:How true was this? (Score:5, Insightful)
I'd say it is pretty hard to dispute. We need mechanisms for identifying someone with say, the early stages of leprosy or birth defects (finding the right mate, etc.).
Parent
Re:How true was this? (Score:5, Interesting)
Many of the examples you've given don't cross over into "uncanny valley" territory - nobody would get stuffed animals confused with real animals...
I thought this too at first but after reading further I realized (at least I think I did) that the PP is talking about real animals that are stuffed after death. So not your cute, cuddly stuffed toy you give to your kid - no one is generally going to be fooled by that.
A dead but stuffed and preserved animal can be difficult to tell apart from its live counterpart, at least until you look long enough and realize it isn't moving.
And contrary to the PP, I do find these disturbing. Just a personal feeling and I do agree that the whole "uncanny valley" thing is a bit exaggerated. But I don't think you can say for sure one way or the other because I believe it's very subjective and there probably isn't a 'one opinion for everyone' take on it.
Parent
Re:How true was this? (Score:4, Funny)
Note: I didn't post a link to Wikipedia out of fear of a picture. [shudder]
Parent
Re:How true was this? (Score:5, Insightful)
Actually, stuffed animals are a good example outside the uncanny valley. Remember, we are talking just about human appearance and action, not animal.
Regarding realistic paintings and statues - yes, they do look like humans, and I would say they reach past the valley on that point, but they don't act like humans. I would also suggest that it is easier for us right now to transcend the valley in appearance than it is to do so in action, if for no other reason because we've had more practice.
Wax figures are another good example. Madame Tussaud's wax figures are excellent representations of the human form - but again, they don't act human, which makes the valley easier to surpass. However, have you ever been to a bad wax museum? I have seen some pretty bad wax figures as well, and they trigger the uneasiness that this theory suggest an inanimate object in this valley would.
As far as cartoons, the physics-defying glass of water - these are all non-human representations, and thus not covered by this theory.
Parent
Re:How true was this? (Score:5, Informative)
I think you misunderstand the uncanny valley concept. In fact, your entire rebuttal is mostly a restatement of the concept itself.
The point is the more realistic something is, the more disturbing any 'defects' in it's simulation are. Stuffed animals don't breath, they don't move, they don't growl. Neither do paintings. These things may be realistic in the sense that they portray a snapshot of the thing they are based on, but they don't come anywhere close actually convincing you that they ARE the thing they were based on. I don't know of anyone who would mistake a stuffed bob cat sitting in someone's den or a museum as the real thing for more than a few seconds. Likewise, people pretty much know when they are looking at a painting or even a photo.
All of those things are on the 'safe' side of the valley. The problem comes when you start getting things that move, sound, and mostly act as if they are alive but clearly aren't. Your actor with the no-spill glass would be in the valley, so to would be photorealistic computer models that didn't have facial expressions when they spoke (ala FF).
And for the record, while the uncanny valley was popularized by talking about computer generated graphics, it was actually coined by a roboticst back in the 70's, and was based on an idea first presented by Freud in the 1910's.
Parent
Re:How true was this? (Score:5, Interesting)
Then there's the opposite. I've looked at a beautiful simple lawn sculpture of a rabbit, thinking I've never seen one that detailed before in my life... then it twitched. And then it was so perfectly still again I thought I might have imagined the twitch. It would let you get within 5 feet before it decided you were too close. It wouldn't even turn its head to keep you in sight.
Oh, and then there's those certain portraits that have depressions on the eyes so that seem to follow you everywhere. Like they're painted on the inside of spoons.
Parent
Re:How true was this? (Score:5, Interesting)
Important to note that the only part of the video that looks completely normal is the behavior of the head, the arms, the breathing of the torso-essentially everything done by the actual human. I'd be surprised if you felt absolutely no (at least confusion if not revulsion) watching the video. The face was interesting to me, but I was blown away by how realistic the hair looked, how realistic the arm movements--until I realized that that was all still a regular human being. Then, when I focused in on the face alone, it simply looked animated, and if not disgusting, at least completely out of place.
I think the general feeling, even if it isn't an all-out feeling of disgust, is one that things are...not...right. Ultimately, I think this is a pretty bad example, though, since Emily is touted as being "not real" but in fact the majority of the body language--the stuff we are tuned into almost subconsciously--is still human. I think if this video skips by any general feeling of revulsion, disgust, or out of place-ness, it is specifically because there are still physical human elements in it.
Parent
Re:How true was this? (Score:5, Insightful)
If the uncanny valley really exist, then please explain realistic paintings that have been around for ages
I think what people mostly call the "uncanny valley" is not the result of a work produced by an artists, but the result up motion captured data applied to a computer model (often generated by 3D scanning). The miss-detection in the data and the incorrect mapping from the animation data to the model result in uncanny results and most often you don't have an artist there to clean things up. When you on the other side have an artists to clean things up, the results most often look quite a fine.
I consider the uncanny valley not something that you drop into the closer you get to realism, but something that you drop into when you screw up the balance between different aspects of a work. A perfectly realistic 3D human will look really uncanny if you just stitch it onto a not so realistic animation, since a lot of vital pieces in facial animation and such would simply be missing. On the other side if you take that same animation and stick it to a simpler human model things look quite fine. Its simply a matter to not move the motion and the graphics so far apart that they won't fit together any more. If you have a super realistic face with every wrinkle modeled perfectly, you better have some animation data to make those wrinkle behave realistically in motion, if you don't you better scale back your detail level, since what looks uncanny is that that is there and looks wrong, not those pieces that are simply missing.
Over the course of the last 20 or so years I have seen a ton of stuff that I would consider uncanny and a ton of stuff that I consider to look quite fine, none of the uncanniness however had much to do with the realism, since even a cartoon creature can look quite uncanny when things are out of balance.
Parent
Not really animation (Score:5, Insightful)
Re:Not really animation (Score:5, Insightful)
All you need is video of somebody of similar build and you can put anybodies face on it.
Parent
Re:Not really animation (Score:4, Funny)
Parent
Re:Not really animation (Score:5, Insightful)
Exactly. This isn't precisely computer animation, it's motion capture minus a lot of steps.
Parent
Re:Not really animation (Score:5, Interesting)
From what i understood, this is simply an easier kind of motion capture that works straight from video without the need for sensors etc. That's not the same as creative animation, you still need a real person talking and moving.
I haven't seen motion capture look this good. But anyway, the point of this is that you could, for example, use a cheap (read: plain) actress for the recording of scenes, and then animate the perfect, beautiful princess character using her mannerisms. Let's face it -- actors get paid the big bucks for their looks, and not so much their talent. There is a hell of lot of acting talent out there that doesn't necessarily have the right "look".
Parent
It's more convincing than... (Score:4, Insightful)
...many flesh-and-blood actors I've seen.
In a discussion elsewhere, someone stated that the facial animation was good, but the body movement was unrealistic. Since the body movement was actually a live actor, I'd say that this was analogous to a passed Turing test -- an observer couldn't tell which parts were animated and which parts were human. (It's a weak analogy, of course, since there was no interaction.)
Re:It's more convincing than... (Score:5, Funny)
Is a Turing test valid if the human is an idiot?
It's ridiculously clear from the video that the face was the "animated" bit.
Parent
Re:It's more convincing than... (Score:4, Insightful)
In a discussion elsewhere, someone stated that the facial animation was good, but the body movement was unrealistic. Since the body movement was actually a live actor,[...]
I noticed, too, that there seemed to be too much weird movement. I actually suspect that's due to the real actress hamming it up too much for the capturing process, trying too hard to make it natural. I bet if they used someone who didn't know what it was for, it would've come out better.
Parent
End of blah blah (Score:5, Insightful)
They failed, and they're lying. (Score:5, Informative)
First off, they failed at getting passed the "uncanny valley". That video is still creepy looking.
Second, this isn't computer animation. It's just video processing. If you still need to do high resolution motion capture to produce your images, you haven't replaced the actor. You've merely edited their appearance in the performance. They didn't even bother to go so far as to take the captured motion and paste key bits of it together into the speech. They just had her sit there and say the whole thing, then "rendered" it.
Lame.
Not a good test (Score:5, Insightful)
Motion capture a face and rerender it from the same viewpoint as a camera used to capture the texture and you'll trivially get something almost indistinguishable from the original. It's only a valid test if you change something significant: move the camera, change the lighting, change the facial features or change the performance.
Re:Not a good test (Score:4, Funny)
I think the measure of the technique is, "how different does Emily look than the actor who supplied the base footage?"
For example, could they make me look like a sixteen year old boy like I portray when I'm chatting?
Oh, wait, I meant to post that anonymously...
Undo, undo, undo!!!
Parent
Wow, quite amazing. (Score:5, Insightful)
Btw, here's a direct link to the video: http://www.youtube.com/watch?v=bLiX5d3rC6o [youtube.com]
Be sure to tick the 'Watch in high quality' when the video opens (anyone knows a way to do that automatically in a link?)
Re:Wow, quite amazing. (Score:4, Informative)
Be sure to tick the 'Watch in high quality' when the video opens (anyone knows a way to do that automatically in a link?)
Add '&fmt=6' after the link. Like this: http://www.youtube.com/watch?v=bLiX5d3rC6o&fmt=6 [youtube.com]
Parent
Yup (Score:5, Insightful)
I'd say it's past the uncanny valley. That's not to say that I can't tell it's fake. She looks a little fake. Something is wrong-- her face is too still or something. But she doesn't look like a zombie. She's not distractingly creepy. That's all they're really shooting for at the moment, right?
Uncanny valley... (Score:4, Insightful)
Uncanny valley in a nutshell: Is it a "Good Robot" or a "Bad Human"?
But, there is an assumption about what is acceptable... what is the norm? At the moment, we're in a rapid transition phase. There are relatively few human-enough-like examples within our day-to-day existence. I would suggest that as these emulants (to coin a term) become more prevalent and pervasive, their familiarity will reduce the perception of their being bad.
We've come a long ways in the 35+ years since I used an ASR-33 Teletype over a 110-baud modem to a time-shared 8KB minicomputer. That sounds like a long time, and in some respects, it is. Today's generation has seen rapid advancements in game consoles, and even now, the best still appear really good, but still unreal. My guess is that in 5-10-20 years, when the visuals become even better, AND THERE HAS BEEN AN EXTENDED PERIOD OF FAMILIARITY, there will be less of a gap to leap. Not just because the visuals got better, but because we have become more familiar with them.
An aside: Look into the eyes of a young baby. Watch how they make eye contact, and don't let go. Watch how intently they examine you. That's setting up neurons and patterns of what is safe, good, bad, and everything else.
P.S. I wonder if the transition from the old black and white TVs to today's HDTV sets has run through a similar perception challenge?
http://www.image-metrics.com/ has another demo (Score:5, Interesting)
And while it's extremely impressive, sadly it's definitely in the valley for me.
The demo is a streaming video ?!? (Score:5, Insightful)
Yay ! Wonderful low-bandwidth youtube streaming video in all its glorious crap-quality !
The best way to show technical demos about photo-realism !
I can't wait to see the thumbnail sized 60%-quality jpeg screen caps, too !
I feel as much informed about the quality as when watching all those wonderful ads about hiddef screens on the TV.
---
Common, Image Metrics, can't you just post a descent hi-quality video file, so we can actually see what your technology looks like ?
Hasn't jumped the valley. (Score:5, Funny)
Heck, she doesn't even look as real as Celine Dion, let alone a real person.
Looks uncanny to ME (Score:4, Interesting)
I wish I'd somehow had a chance to view this before knowing that it was a computer animation... say, a side-by-side comparison of a real and an animated person and a challenge to guess which was animated.
To me, "Emily" did not look real and did look uncanny. Actually, it reminded me of nothing so much as one of those videos where they replace a baby's mouth with animation so that it appears to be talking like an adult. It seemed to me that the animation's "mouth" was not stably positioned on its "face;" when the head turned, I perceived a change in the position of the mouth relative to the face. Something about the skin didn't look right, either.
Would I have accepted it as real if I were expecting "real?" Yes. But that's not the same thing.
Some years back I took part in an experiment to gauge something about necessary bit rates and algorithms to make synthesized speech sound real. What struck me forcibly was that, in this experiment, when you were listening to the best synthesized speech, if I'd had no standard of comparison I'd have said it was real. But when they switched to a real voice saying the same thing, there was the most amazing sensation, almost a tactile sensation of sound shaped by warmth and moisture. Only after you heard the real thing did the synthesized speech seem cold and mechanical.
It's "performance transfer technology" (Score:4, Interesting)
Image Metrics calls this "performance transfer technology". It's not really animation; it's more of a scheme for pasting face A onto actor B. Quite a bit of this already goes on; often, when you see a stunt performer's face on screen, the face of the principal has been transferred to the image of the stunt performer. With this new technology, that can be done without matching camera angles or going through the whole "dots on the face" makeup ordeal.
Introducing Emily O'Brien the real actress (Score:4, Informative)
Am I missing something? (Score:4, Insightful)
So they make a 3D model of Emily's face (using a 3D scanner, presumably), then they film Emily moving her face, then they deform the model to match Emily's facial expressions, then they superimpose the model on Emily's head.
Er... what for?
At best they'll end up with something identical to the original (but they don't - the model doesn't wrinkle properly and sometimes the tracking is slightly off - you can see her face "float" relative to the hairline and ears).
I could understand the point if they could take expressions from one person's face and replicate them on another person's face (which is something you can do with motion capture - and some clean-up work). But obviously they can't do that automatically, or they would have done it for the demo.
I can see this kind of technology being useful to disguise the transition between an actor's real face and a 3D face (which will later be deformed by hand, or morphed into some creature, etc.), but the demo is so limited (camera doesn't move, the 3D face is almost identical to the real face, etc.) that it seems a long way off from being an alternative to motion capture and manual tweaking. This is like showing some (supposedly) revolutionary new GPU by making it print "Hello World" on the screen. If the technology is so great, why such a limited demo?
Wait until eye contact w/robots becomes reality (Score:5, Interesting)
An even bigger problem will be making robots that can convincingly pass for human while physically in their presence and trying to feign one-on-one communication. Have you ever noticed that somehow, something just kind of clicks, and you *know* you've made eye contact with someone... and you know that THEY know, too? They might be far away, in a moving vehicle, looking at something else (or just generally looking around), but every now and then "it" happens... you make random, fleeting eye contact with a stranger.
My theory is that it's due to the fact that your eyes are always moving (if your eye were perfectly still, you wouldn't be able to see, because rods and cones derive most of their information from CHANGES rather than instantaneous sampled state). I'm guessing that the pattern of movement appears random, but somehow the part of your brain responsible for background signal processing is able to recognize that movement pattern in the eyes others, and tries to synchronize itself to it. Neither person is intentionally trying to do it, or is even aware of it, but their brains -- through visible eye movement -- are actively negotiating the equivalent of a handshake... and when it happens, a metaphorical "datagram" gets sent to your conscious brain letting you know that you've "locked on" to another person. When you're intentionally talking to someone, it lets you know that you have their attention. When it unexpectedly happens at some random moment when you're just gazing out at the horizon, it can be awkward and uncomfortable.
It's why if you're trying to hide, the worst thing you can possibly do is try to watch what's going on nearby. You might be in the dark shadows, or behind a large object with little more than a hole big enough to see through... but somehow, if someone happens to gaze in the right direction, and their eye detects the movement pattern of an eye somewhere nearby, they're going to immediately feel like something is amiss, even if they don't immediately realize what just happened. If their gaze crosses the gaze of another person who's looking at something entirely different, it might just be a feeling of unease. It's why looking for a lost person or animal is easier than looking for a lost object, at least if you're close enough to potentially make eye contact, Looking for a misplaced object, your brain has to process everything it sees, and constantly do pattern-matching. With people and animals, it's kind of like they're emitting a short-range beacon that allows you to randomly gaze around, but get "that feeling" whenever eye contact occurs, signaling that some area merits further visual inspection.
Anyway, getting back to the Uncanny Valley, it'll be interesting to see what impact the ability to feign eye contact by robots will have. A robot with no eye contact seems creepy in a "dead" kind of way. Would a robot that "almost" managed to maintain eye contact be MORE comforting, or creepier still? Would the "comfort" factor depend upon whether the person interacting with the robot KNEW they were interacting with a robot? Or would making "almost correct" subconscious eye contact with a robot send chills down the person's spine, setting off subconscious alarms to let them know, "DANGER! Something here isn't quite right!", regardless of whether the person KNEW it was a robot?
Re: (Score:3, Funny)
Re:Uncanny mask (Score:4, Funny)
So gluing an weird uncanny mask on an actors face will be the future of animation?
Considering the quality of acting these days by Hollywood, anything that obstructs their faces would be an improvement.
Parent
Re:It's very close. (Score:5, Informative)
That's what I thought as well until I saw it in higher quality.
A higher quality version of the video can be found here [awntv.com].
It's not perfect, but it certainly is climbing high up out of the uncanny valley to say the least.
Parent
Re:It's very close. (Score:4, Interesting)
I noticed the same thing, but I really don't think it would have "jumped out" at me if I hadn't been looking out for something fake - I'd just have assumed that "Emily" had strange expressions.
I think Small Furry Creature [slashdot.org] is on the right track - the uncannyness of the valley isn't "people" looking almost-but-not-quite-right, it's our physics assumptions failing - when fat on someone's face doesn't move the right way, hair doesn't fall the way we expect it to, and so on. They got around that in this video by using real background video everywhere except for the face, so there are fewer cues for us to notice physics going wrong - except, as you point out, the eyelids. That's how they got around the uncanny valley, imnsho.
Parent
Re:It's very close. (Score:5, Interesting)
Where's her tongue?
Maybe it's the grainy movie, but as a lip-reader, it caught my eye... that, and many face "wrinkles" that usually appear when smiling, frowning, or even raising your eyebrows seems to be missing as well...
Parent