On Game AI In The Uncanny Valley 87
An anonymous reader writes "Normally, the Uncanny Valley theory is used to critique graphical realism in games, but it also applies to AI. Therefore, designer David Hayward examines AI's Uncanny Valley over at Gamasutra, citing games from Valve's Half-Life 2 to the interactive drama Façade." From the article: 'There's a small minority of people who are consistently strange in particular ways... I don't mean to pick on them as a group; nearly all of us dip into such behavior sometimes, perhaps when we're upset, out of sorts, or drunk. Relative and variable as our social skills are, AI is nowhere near such a sophisticated level of interactive ability. It is, however, robotic. Monstrous and sometimes unintentionally comedic; the intersection of broken AI and spooky people is coming.'"
For those that haven't seen it... (Score:5, Funny)
Re: (Score:2)
Anyway, I don't think the uncanny valley will be a factor in AI for quite some time. There's just far too much to overcome.
Re: (Score:1)
Re: (Score:1)
Uncanny valley... adaptive levels? (Score:5, Insightful)
As our technology improves to create better and better artificial representations, our ability to detect them does as well.
Re: (Score:3, Interesting)
And btw: There have been studies comp
Re:Uncanny valley... adaptive levels? (Score:4, Insightful)
The composite faces you mention are likewise on the near side of the valley. Images of faces fabricated from scratch often are likewise, with current technology, due to the fact that still images contain so much less information than moving ones. Compare modern CGI faces to those of the late 90s and you'll see how they gradually got better, climbing the near slope of the valley and becoming more believable and identify-with-able.
Re: (Score:2)
The Uncanny Valley is an conclusion, not an observation and doesn't even apply to computer animation, it was originally meant for robotics. The observation is simply that things can look uncanny, I don't doubt that one. Where I have a problem with is with the claim that they will look more uncanny just because they are more realistic or as some interpret it that more technology will mean worse looking graphics due to t
The Uncanny Valley in two sentences (Score:4, Informative)
I still call bull (Score:5, Insightful)
E.g., let's take two sets of models which were both in the "uncanny valley" if it exists. On one end you have the extremely detailed models of Final Fantasy: The Spirits Within, and on the other the models of EQ2 which a _lot_ of people described as "lifeless", "sterile" or other such euphemisms. I'm one of those. In fact I'd call them just disturbingly wrong, the kind where your subconscious keeps snapping out of suspension of disbelief screaming "that's not a tree!" and "that's not a human!" If an uncanny valley exists, then they're in the uncanny valley too, right?
Then something that falls in between those two points should be in the uncanny valley too, right?
Well, wrong. Oblivion for example was a lot easier to swallow than either, although detail-level-wise it's between the two.
That's called a Reductio ad absurdum [wikipedia.org] proof, where assuming X leads to the false conclusion Y.
Furthermore, if you've actually read the Uncanny Valley theory, the examples used are blatantly bogus hand-waving. E.g., yes, a zombie is disturbing, but it's bogus to claim that it's purely for aesthetic or "how much it resembles humans" reasons. There's a whole bunch of cultural and emotional meaning tied to that, and claiming that it trips people's fears just because of the "uncanny valley" effect, is like claiming that you fear a car coming at you just because the headlights look sorta uncanny like eyes.
And again, you can do a reductio ad absurdum there. The Undead in WoW are the most disturbing visually, the characters in Spirits Within are uncanny valley too, so something in between should be in the Uncanny Valley too. Yet the Wow humans and elves are considered the races that look good.
In fact the zombies are the perfect counter-example all by themselves. If lowering realism moves something out of the uncanny valley (e.g., lower polycount characters in games are less disturbing than the characters in The Spirits Within), then it should do the same for zombies. It doesn't. They're still repulsive even at Quake 1 polycounts.
Or if slight imperfections like in Spirits Within, or details like teeth on a vampire, are what makes them disturbing via an Uncanny Valley effect... then how about pointy ears on elves? Shouldn't Legolas cause the same effect? Well, bummer, he doesn't.
Etc, etc, etc.
The Uncanny Valley is one of those things that makes sense only as long as you don't actually use your brains.
Yes, there are all sorts of ways in which being different or acting not-right can trip people's suspension of disbelief. That much is obvious. But there is no single dimension measuring it, and no single Uncanny Valley graph. There are thousands of factors which can be right or awfully wrong or somewhere in between, and thousands of fears, beliefs, expectations that can be tripped by it. You can't take the average and use it as the X axis for an uncanny valley graph, because even if the average is 99% right (hence the whole should be on the right side of the valley), one single detail (e.g., "omg, they're zombies") which can be disturbing on its own.
E.g., The Spirits Within wasn't just "a littel off", it had outright bad acting. That's what tripped people's suspension of disbelief. There was no uncanny valley effect, no overall being just a little off, it's just what you'd get with human actors acting badly.
It also overlooks the problem of expectations. The Spirits Within is wrong because you expect them to be human, Toy Story or Oblivion aren't because you expect them to be respectively toys or NPCs in a computer game. You have different sets of expectations for them.
Aesop's Fables (since they keep getting mentioned as Uncanny Valley effect exam
Re: (Score:2, Interesting)
There is no one-dimension axis measuring likeness to human.
It almost certainly differs from person to person, and we may not be able to define it yet in words or symbols, but there must be such an axis for the simple reason that we can define one, and experimentally determine subjects' positions on said axis. A simple test asking a large, random sample of people to rate a number of subjects on a scale of 'lifeless cube'=0 to 'pure believable human'=100 should give you enough data points to get a rough 'normalized humanness' axis. Ask them to crossrate how much the
Oh please... (Score:3, Insightful)
The logic I'm proposing is along the lines of:
- _if_ there's such an axis and the curve looks like that (i.e., with a single dip below zero), then the "uncanny valley" zone is basically an interval
- if two points are part of the same interval, then a point in the middle is part of that interval too. (E.g., taking the in
Re: (Score:2)
It also means that if you're having problems with yo
Re: (Score:1)
Let's look at a couple of examples that were mentioned:
FF: The Spirits Within: Very realistic character models in a general sense. The faces of the characters were very waxy, with few lines or minor muscle move
Re: (Score:2)
Anyway, a low resolution painting can look vary human and can easily cross the valley as IMO did stills from the spirits within. However, moving 3d avatars bring a new level where they way things move and style of movement is more important than the number of polygons. The way cloth moves, how people look at things
Re: (Score:2)
As it is, there are plenty of almost-real characters in games, and they fit people's expectations fine. Consider Final Fantasy 12. Nobody can pretend the characters look real, but they don't trigger the sense of rejection that the Spirits Within did. And if you think that's perfectly-real (which it isn't, by a long shot),
Re: (Score:1)
a zombie is disturbing, but it's bogus to claim that it's purely for aesthetic or "how much it resembles humans" reasons. There's a whole bunch of cultural and emotional meaning tied to that
But the cultural meaning of zombies is of very recent inception. The concept of modern zombies is just a corruption of certain Vodoun beliefs and the real popularity of the idea really only just arose with the film industry. There were prior beliefs in some forms of the undead such as the Middle Ages' beliefs in revenants but they were thought of as emaciated corpses or skeletons. Nothing as almost-human as a zombie. I'm cribbing the hell out of the zombie wiki here, but the point is the cultural associati
Re: (Score:1)
Re: (Score:2)
The Uncanny Valley Hypothesis (Score:2)
This can be easily observed. The unrealistic humans in "the incredibles" seem much more human than the children from "Polar Express," even though Polar Express uses a much more realistic rendering style.
Wikiped [wikipedia.org]
Re: (Score:2)
Polar Express looks crap, because they have done a bad job, not because its more realistic. Final Fantasy: Spirits Within looked better, so did Advent Children and Gollum even more so. I mean what do you expect if you use crappy looking 3D models and then map the motions of a 50 year old guy to 8 year ol
Re: (Score:2)
I still think you're missing the point of the hypothesis. Read the wikipedia article.
Re: (Score:2)
Gollum looks quite realistic to me, fantasy creature sure, but other then that they did everything they can to make him as real as possible. Some old DonkeyKongCountry cartoons look a lot more uncanny to me, even so they are a lot more cartoon then Gollum or Advent Children ever was. And lets not forget that between Advent Children and Spirits Within you had five years of tec
Re: (Score:2)
Again, you don't undestand the hypothesis. Please read up on it. Thanks.
Re: (Score:2)
Re: (Score:2)
First, yes, you're right, voice and writing and other factors play a role, but this is totally irrelevant as the hypothesis does not concern itself with voice and writing and other factors. It only looks at how realistic the character is rendered. I don't understand why you keep bringing up all these other points. Yes, they exist, but it is utterly nonsensical to try to disprove the Uncanny Valley by bringing up things that have nothing to do with the hypothesis. It's as if
Re: (Score:2)
The original hypothesis doesn't even concern about computer animation, let alone games, its about robotics and nothing else. When speaking about games and renderings story and writing do matter a lot. You can use stick figures and wireframe, if the voice acting and story are interesting, nobody will care. If story or
Re: (Score:2)
Let me say this again: The Uncanny Valley Hypothesis only concerns itself with the influence of realism on the empathy felt by an observer. It does not matter that other things also influence the empathy felt by an observer. Obviously, a well-written story will make the observer feel more empathy, but that doesn't matter for this hypothesis. Why in the world are you incapable of understanding that?
And again, this is not a scientific theory. It's a hypothesis. Obviously you can't measure how human something
Re: (Score:2)
Somebody builds a zombie looking robots and concludes that all robots have to look like zombies. Look for example at:
http://en.wikipedia.org/wiki/Image:Repliee_Q2.jpg [wikipedia.org]
Why does it look wrong? Some obvious reason would be the arm and hand positions, a human doesn't hold his fingers like that, heck, I can't even hold my finger like that if I would want t
Re: (Score:2)
Re: (Score:2)
No, the thesis doesn't state that things get harder when you get closer to realism. It states that things will look uncanny the closer you get to realism till you hit a point where things will go better again, very different thing. If things just would get harder there never would be a valley to begin with, since you could always jus
Re: (Score:2)
No, the thesis doesn't state that things get harder when you get closer to realism. It states that things will look uncanny the closer you get to realism till you hit a point where things will go better again, very different thing.
GAH! If you fix the "small mistakes,"* obviously it will become more realistic, thus getting out of the valley again. IT'S THE SAME THING! Obviously you're just arguing for argument's sake, and you're not going to admit that at this point of the discussion, so I will end here and just encourage you to - if you indeed still don't get it - read through the wikipedia article again. I'm sure you'll figure it out. Good luck.
* the mistakes, by the way, may look small, but that doesn't mean they're easy to fix.
Re: (Score:2)
Lets make this short, as I said you *CAN'T* cram something as complicated as computer graphics on a single "cartoon - realism" axis. There are not just real graphics done badly, there are also things like more simplistic graphics done well, you simply can't compare those when you ignore what makes one good and the other bad.
It's freaky (Score:2)
Playing with bots and people, REALLY unpleasant.
Re: (Score:3, Funny)
Re: (Score:2)
Re: (Score:1, Insightful)
for example, i know if i touch a hot flame, it will burn me and it will hurt. i may not want to touch the flame, but i can decide to anyway, even against my own will/judgement/past experience and every other possible variable that can be taken into account. AI is almost the opposite. it's static. even if you were to throw in a "random" variable that ends up making the final decision between any number
Re: (Score:2)
* Though in some sense the only invariant in the human brain is physics, it's probabl
Re: (Score:2)
Re: (Score:2)
I would assume that sophisticated game AI algorithms wouldn't use boolean logic as much as some kind of "fuzzy" logic. That is, it would take into account the number, strength and proximity of known enemies, adding up to some "threat level", compare that to its own "capability level" which includes things like hea
Re: (Score:2)
Re: (Score:1)
I think you mean a feedforward [wikipedia.org] neural network instead of a Kohonen [wikipedia.org] one. And even then you're overestimating how smart learning algorithms will be compared to engineered ones like state machines [wikipedia.org] and production rules [wikipedia.org].
You're right about the predictability and scripting though. This is exactly to avoid the uncanny valley. Automatic AI keeps them dumb, but fun to play against/with and when they need to behave more like humans, a human inputs exactly what they should do or say.
Interesting article (Score:4, Interesting)
I disagree with this. I think in the future, game programmers won't have to go as far as the psychology departments of their nearest schools. They'll just have to walk over to the nearest cubicle and talk to the animators working on the game. As game models have become more and more complex, companies are using more and more motion capture to capture action sequences, but animators (especially good ones) are trained to make a non-living 3d model imitate human behavior. There's over 50 years of research done for animation by animators on how to bring life to drawings and 3d models in motion - which is something that can be directly transferred over into programming terms as opposed to a research paper on a psychological disorder. An animator can tell you when to make a character blink in order for it to appear more realistic; a psychologist, not so much.
btw, IAAA (I am an animator) so I'm slightly biased
Re: (Score:2)
Animators know what makes dead 2-D filled shapes feel "human" and likable; if they applied their skills to real people, however, they would make them look dumb, hacked, and repulsive.
Just imagine Bill Gates Wile-E.-Coyote-ing off a cliff, or Angelina Jolie with Japnime wrinkles and random airlines, buldged eyes, and blue fluffed hair... No ma
Re: (Score:3, Interesting)
A real animator knows what makes a human human-like and given enough time will be able to fix any issues that arise. The issue isn't with animators, its with automated systems like motion capture, since no matter how many reflective dots you glue on an actor, they will always be off by a little bit and that is what gives yo
Re: (Score:2)
Re:Interesting article (Score:4, Interesting)
So, in response to the quote (I suspect that consultation with and evaluation by psychology departments may become relevant to game AI in the coming years, given that they're the most comprehensive resource in existence on human behavior), yes, psychology will play a large role as we are able to dedicate more computing resources to AI. This isn't conjecture; I and many other researchers are already consulting with psychologists in designing games.
Re: (Score:1)
Procedural animation (Score:3, Interesting)
I keep thinking that procedural animation [wikipedia.org] is going to be the next big thing. Instead of rigid animations, we'll see rules governing the position of each limb and how they interact with the world and other characters. It's expensive in terms of CPU processing power, but it allows for a far more natural interaction with the environment. There's
Re:Interesting article (Score:5, Insightful)
Re: (Score:1)
But how do we calculate the emotion? No matter how complex the representation, abstractions are made and details are left out. And the details are what make AI lacking.
TFA is short but its an excellent commentary on the fundamental issues.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
http://www.animationmeat.com/pdf/nineoldmen/ELEnt1 _HamLuske.pdf [animationmeat.com]
http://www.animationmeat.com/pdf/nineoldmen/ELEnt3 _Acting.pdf [animationmeat.com]
http://www.animationmeat.com/pdf/nineoldmen/ELEnt5 _Action.pdf [animationmeat.com]
http://www.animationmeat.com/pdf/nineoldmen/ELEnt1 3_Personality.pdf [animationmeat.com]
http://www.animationmeat.com/pdf/nineoldmen/ELEnt1 4_BeYourself.pdf [animationmeat.com]
if you're interested in the rest of the articles, you can check them out here:
http://www.animationmeat.com/notes/nineoldmen/nine oldmen.html [animationmeat.com]
"We should not
Re: (Score:2)
Humans may not fair much better. (Score:3, Insightful)
But what kind of AI . . . (Score:1)
Re: (Score:2)
Re: (Score:1)
Well, if I want a "game" that's intended to be a surrogate for social interaction, then I'd want emotionally responsive AI. Or an MMO.
The AI you put into a video game or a vaccum cleaner should not be the same as the AI you put into something that's going to assist the sick or elderly.
It is cute, I don't like, I like ... that valley? (Score:2)
Game theory and Complex Systems mathematics expressed 3/4D visually will greatly reduce wait time for virtually natural (complex and non-repulsive) AI/Robotics and human collaboration/relationships. From the familiar to the revolting to acceptance we will go over the next (maybe) 30 years.
http://necsi.org/ [necsi.org]
http://www.complex-systems.com/ [complex-systems.com]
http://en.wikipedia.org/wiki/Complex_adaptive_syst em [wikipedia.org]
Presently the biggest pr
It has already arrived: (Score:4, Funny)
Ever "talk" with Eliza?
That's a broken personality right there...
Re: (Score:3, Funny)
-
Have you ever BEEN Eliza? (Score:3, Interesting)
Are you stuck in a conversation with a zealot ranting about how BSD is better than Linux, or why Gentoo sucks compared to the Gay Penguin distro? Have to listen to someone ranting about what subtle differences make Manowar the greatest band ever, and Metalica just a bunch of soulless sell-outs? Have to nod through as your GF goes on at great lengths as to what should
Re: (Score:2)
BTW, which Linux distro is gay?
The Author Missed the point (Score:3, Insightful)
The author is missing the point of the uncanny valley. The point of the uncanny valley is the dip. If you have a non-realistic pixar-like character (The Incredibles), you empathise with them. If you have a more realistic, but not good enough character, the character gets creepier (Polar Express). Then, when it gets really good, you can empathise with them again (Hugo Weaving, i.e. Agent Smith, in the Matrix Sequels).
The author basically says "AI is hard". But he doesn't make any real argument as to there being some "valley" where as the characters get less realistic they act more believably, and as they get more realistic, they also act more believably. A much more accurate title would be "AI is a steep hill".
Re: (Score:3, Insightful)
I find many cutting edge AIs unnerving to interact with because when they do the w
Re: (Score:1)
Real AI a long way off. (Score:2)
Re: (Score:1)
The game I've played most over the last few years is Diablo II LOD, and there is a reason why the mercenary is usually called the moron. In addition, I wonder why the term role-playing game is used to describe that game, and a few more, when there is no ability to role-play at all. I
Re: (Score:2)
This is actually one of those topics I've written a lot about various places (think there's an ancient blog of mine somewhere with a pretty extensive discussion of the future of AI). There's a couple of things going on here. The first is that "artificial intelligence" really isn't a good name for it because it gives people unreasonable expectations of human-like behaviour. Think about it. If gnuchess were truly artificially intelligent, wouldn't it make more realistic mistakes?
I think people hear "AI" and
Re: (Score:2)
I think we're a long way off from seeing computer controlled characters that aren't merely following scripts
http://www.amazon.com/Scripts-People-Live-Transac
Re: (Score:2)
What about talking AI too? (Score:1)
Well if you didn't already know... (Score:1)
The Uncanny Valley is a hypothesis about robotics concerning the emotional response of humans to robots and other non-human entities. It was introduced by Japanese roboticist Masahiro Mori in 1970, although draws heavily on Ernst Jentsch's concept of "the uncanny," identified in a 1906 essay, "On the Psychology of the Uncanny." Mori's hypothesis states that as a robot is made more humanlike in its appearance and motion, the emotional response from a human being to the robot will become increasingly positive and empathic, until a point is reached beyond which the response quickly becomes that of strong repulsion. However, as the appearance and motion continue to become less distinguishable from a human being's, the emotional response becomes positive once more and approaches human-human empathy levels.
The wiki [wikipedia.org] has a picture which illustrates the name. Now you can go back to arguing esoteric robot hypotheses.
But even if AI... (Score:1)
some simple tricks make it all seem real (Score:1)