Wrestlevania

"There is no uncanny valley any more," declares French developer

Recommended Posts

Fucking hell! That's not even remotely funny Wrestlevania.

I don't know. I laughed out loud.

Share this post


Link to post
Share on other sites

I think u're being too harsh on the makers of heavy rain, have u seen the audition tape of the woman who the character is based on, she looks like she’s from the uncanny valley, so they either did a really good job in making the film and a terrible one in casting or did a brilliant job in casting (by finding the most unrealistic person ever) that covers up their inability to make a realistic person.

I thought the voice was either computer generated or done by a deaffo, but it turns out she is not a native of an English speaking country which is why the voice goes a bit mental in bits (getting a non English speaking person to speak English adds weight to the case that they cast her because she was the most unrealistic person they could find.) or could be as they made it France or something.

Check it out here

Share this post


Link to post
Share on other sites

I believe Quantic Dream are in fact French.

But it's an interesting point you make all the same, about masking their short-comings by using "imperfect" source material in the first place. Might just be that she's just a good voice actress though.

Edit: It's a little disappointing to see her face remain almost expressionless, even though she's delivering those lines with some panache.

Share this post


Link to post
Share on other sites

Looking at that original footage, the motion capture simply destroys her acting. The scary face posted above doesn't look nearly as freaky in the original, even though her face is full of weird shit for the motion capture.

Share this post


Link to post
Share on other sites

At that point her real face is very exaggerated and she lifts her eyebrows up loads, the translation into the virtual model is worst, partly largely because the white of her eyes is too white.

I agree with wrestle, she does deliver some of her lines really nicely (despite the spacko bits where it all goes a bit mammaerrmam) but she doesn't know what to do with her face while delivering those lines so completely overacts and accentuates all her facial motions, then the motion capture exaggerates it even more.

Share this post


Link to post
Share on other sites

I agree somewhat with the 'heavy rain was horrible', on the other hand I clearly remember this video - despite it being amongst tons of other videos I watched on the net since that was released. Whatever it was - it seems to me it at least was memorable... (maybe its just the uncanny valley - to even reach the uncanny valley should be regarded as an accomplishment - they should use that to make a horror game, where everyone is human .. but .. not .. quite ... dum dum duuuum .....)

Share this post


Link to post
Share on other sites

The problem was never whether something can be distinguished as CG or not, was it? I thought the problem was people acting like some kind of glitchy replicants.

Share this post


Link to post
Share on other sites

That's not what the uncanny valley is at all =( Glitchy replicants would actually be unvalley uncanny.

Share this post


Link to post
Share on other sites

This graph from the Wikipedia article on the uncanny valley is so awesome:

461px-Mori_Uncanny_Valley.svg.png

Share this post


Link to post
Share on other sites

Hehe that is pretty cool indeed. It also helps the discussion. You see, good-looking CGI that doesn't mind looking like CGI would be at the peak immediately before the dip into the valley (eg The Incredibles). The valley occurs as the CGI gets closer to "real" life. But it has to be travelled through before reaching the grail of "healthy person". Quantic dreams claims don't hold up because they're not even into the valley yet.

Where's Joanna, isn't she meant to be doing a thesis on this?

Share this post


Link to post
Share on other sites

Nice graph. But what are bunraku puppets doing that high on the better side of the valley? :erm:

Share this post


Link to post
Share on other sites

I wouldn't know a bunraku puppet if someone made it pinch my arse in the street, but it's worth pointing out that the examples given in the chart above are all applicable to the field in which the hypothesis originates: cybernetics. It was first observed in Japan in the 1970's, so CGI is just a much later field in which it also seems to be applicable.

brkl, the UV is through and through a hypothesis, not fact. Nonetheless, adverse reactions are observed to many different near-real representations of other humans, and this doesn't tend to occur with representations of other species or objects. To take a couple of counter examples though, vets and geologists might disagree - what fools you or I won't convince a specialist so well, and due to the quirks of us being a social species, we're all specialists in other humans.

The closest anyone has come to explaining the why of it is that near-real representations trigger an adverse response originally developed to make humans keep their distance from ill people. This idea is interesting, but pretty damn far from proven though.

The negative reaction also has subjective boundaries and lifespan. The reaction isn't uniform from one person to the next, and even if someone has it, they may get used to it during the course of a film. Polar Express makes my skin crawl to this day, but I soon got used to the characters in the Final Fantasy films. Others have reacted very differently.

Share this post


Link to post
Share on other sites

So, I'm right? It's not about whether something is clearly CGI or not. Any CGI doesn't make me feel uncomfortable, this :scary: does.

It's not really even about graphics. Even if we had a 100% accurate graphical representation that could speak a line or two convincingly, faulty AI could make us feel like we're talking to a person with only half a mind or flashes of sanity amidst machine-like behaviour.

Share this post


Link to post
Share on other sites

You're not quite right; the only thing that matters is how lifelike something is, so if something is clearly CGI, it steps away from that feature, through the uncanny valley and back into the realm of the non-realistic. So it does matter whether something looks like CGI or not.

Share this post


Link to post
Share on other sites

I agree that CGI which still clearly looks CGI hasn't passed the uncanny valley yet even if it doesn't creep you out at all. I'm not even sure if the Heavy Rain (The Origami Killer ?) casting video really reached the bottom of uncanny valley. It might have been terrifying for other reasons.

Share this post


Link to post
Share on other sites

Whether you're right or not, I think it's fair to say they've done a fucking good job of those eyes.

Share this post


Link to post
Share on other sites

That looked pretty cool, but it was nothing more than a mo-cap of the face. I wonder why the eyes were so expressionless all along... If I were to try to demonstrate how awesome my expression-capping system was, I'd have someone make all sorts of expressions, not just starte blankly back and forth.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now