Video: Tokyo researchers build facial reactions into cat head
Posted by GreenReaper (Laurence Parry) on Tue 15 Nov 2011 - 16:47 — Edited as of Thu 24 Nov 2011 - 21:10
Researchers at Tokyo Metropolitan University have combined non-contact motion detection with animatronics to build a fursuit head which reacts to a performer's facial movements.
The technology, exhibited at the 19th International Collegiate Virtual Reality Contest, allows the wearer to control the eyes and mouth; ear and eyebrow control is planned. [scottbob3]
About the author
GreenReaper (Laurence Parry) — read stories — contact (login required)a developer, editor and Kai Norn from London, United Kingdom, interested in wikis and computers
Small fuzzy creature who likes cheese & carrots. Founder of WikiFur, lead admin of Inkbunny, and Editor-in-Chief of Flayrah.
Comments
No... thins is wrong.
Uncanny valley much?
It's not about being realistic/unrealistic; it is about knowing the boundaries where imagination ends and unhealthy obsession begins.
FURRIES BELONG IN FICTION, NOT IN REALITY.
You're learning. Smart enough to see uncanny valley ain't the end all/be all of creepiness.
But it's just a special effect for future Disney theme parks, really. If even that. The eyeballs blink and the mouth moves. The technology itself is pretty impressive, if you get what they're doing, but for a fairly simple end result.
When they get to the "eyebrows and ears" part, it'll be slightly more impressive.
Though I'm a bit confused about the moving ear thing. How's that going to work?
How it is going to work:
http://www.youtube.com/watch?v=RSf9aEETnvE
In the end, it's just a primitive robot suit.
In fact, if you are the type who wears robot suits, why even bother with the fursuit?
Like spaceships, pad-format computers and 3D holograms?
One day, furries will be a reality. The stories written today will shape our treatment of them in tomorrow's world.
Man, I sure remember all those great sci-fi pad-format computer stories.
Commenting on your smart phone is the pits, even if it is in context apt.
About furries in real life; the main problem is where's the profit? It would take a lot of money to make "real" furries at the level the fandom desires. A pig with human genes for easier heart transplants? Sure, you could get funding for that. A two-legged fox/human hybrid? Sounds neat and all, but what would you do with it?
Obligatory Jurassic Park reference (it's the newer, sexier Frankenstein!): Michael Crichton made his dinosaurs theme park attractions for a reason. Without a way to sell the dinosaurs, there was no real need to make the dinosaurs. Scientific research today is largely funded by corporations looking to make a profit; there is funding available for science for science's sake, but not on the "let's make furries, yay!" level. Hell, it wouldn't even arguably be science for science's sake; theoretically, we already know how it can be done. Going ahead and doing it would not even teach us anything we didn't already know.
People ranting in their basements that the "scientific community laughed at my genius, but they won't laugh when I show them my creation!" are often a. living in their parent's basement and b. even if they had the actual talent to pull a Tony-Stark-creating-Iron-Man-with-scraps-in-a-cave level feat in the even harder field of biology, they don't have all the time in the world to do it because if they call in sick for another shift at Pizza Hut they'll be fired (the manager sympathizes, I mean, he used to be a real reporter, after all, times are tough, but seriously dude, he's sick of covering for you) and that means no more money for the lab rats who would have become sexy furry superheroes if he just HAD MORE DAMN TIME AND MONEY! So, anyway, wouldn't put much faith in the lone rogue mad scientist method. Besides, the sexy furry superhero rats would turn evil in the end.
The theme park animal people thing would be ethically wonky, and everyone would say you guys just stole the idea from Jurassic Park. Cheap labor force? No, because it would be the most expensive cheap labor force ever, and if it's not like we don't need to create a group of people to treat unethically in that way. Let's see, sex slaves for rich perverts... God, let's hope not. In this case, it's not whether or not we should, or whether or not we can, but whether or not we should bother.
Some people just can't wrap their heads around the idea that turning into a different creature by means of science/magic/magitek will NOT make their lives better.
Others can't wrap their head around the idea, that while many people have emotional issues that would not be solved by being able to change their body, that doesn't preclude it from being a net positive for others. Even if it is just a matter of fun.
Speaking as a (computer engineering) researcher: We do things because we think they're *awesome*. Finding an excuse that will fly with grant agency is a consideration, but quite a lot of creativity is usually brought to bear on that. In this case, the techniques that would give you customizable furry-like humans could be used to make just about any other morphological customizations you wanted. That would be worth quite a lot of money indeed (be stronger, faster, slimmer, more attractively-shaped, etc), though it would likely just affect your future kids, rather than you.
On the uplifted-animals front, the Moreaux series gave a very plausible motivation for genetic engineering: spies and cannon fodder. The pathfinding project was for normal-ish guard dogs bright enough to perform human-level observation and judgement. Follow-on projects were for all manner of bipedal animal-critters, used for the same reason drones are being used in the real world: losing human lives in a war is politically expensive. Losing hardware (or second-class nonhumans) is much less so. The development program is expensive. Replication much less so (and drones are pretty damned expensive too - "Morey" soldiers would effectively be drones taking an infantry role).
With regards to one of the trolling comments later in the thread, the fact that people will want this due to poorly-thought-out motivations is actually a _selling_ point. The entire advertising industry is based around getting people to make impulse buys and poorly-considered decisions. If any industry group thinks that they can pitch "become a cat-girl!" as stylish - or any variant of similar technology - funding would be obtainable.
With regards to already knowing how to do this - no. We don't. Not even close. There's a ridiculous amount of complexity in the systems that determine _how_ an organism grows from a single fertilized cell to its final shape, and even greater complexity in how all of the disparate subsystems interact. Animals are not lego kits. Engineering a critter that would grow up to be a furry-like human/animal hybrid, and live, is highly non-trivial. Per above, the techniques used for that would have wide applicability elsewhere (and are being studied, for other purposes; there's just a long way to go).
... so... much... text...
What, you can't make it through four paragraphs? You never read a *book* in your life?
Step away from the computer, dude, and get an attention span. . .
I am reading Anna Karenina right now.
But here is the problem: if you take more than needed to explain the point, then the reader gets bored. What was written could be re-written in the form of ONE SHORT PARAGRAPH.
There is a balancing act between fleshing out points and risking people not reading it. Although probably better in the end to have a small minority of one or two people "tl;dr" than to have many just assume one is an idiot for lack of content and depth.
>although probably better in the end to have a small minority
>of one or two people "tl;dr" than to have many just assume
>one is an idiot for lack of content and depth
Has anyone really been far even as decided to use even go want to do look more like?
Mister Twister, you're babbling.
No, I just did not ge
There were four points.
One point per paragraph.
Short enough for you?
WATever ¯\(o_°)/¯
The problem here is that is both expensive economically and ethically; doing something "just for fun" that is both prohibitively expensive as well as likely to get you bad publicity is the problem. This is why I truly doubt we'll see animal-to-human(like) transformations.
I've already answered the cannon fodder question; cannon fodder shouldn't be expensive. Makes an interesting sci-fi series; that's about it. Going the other way, if a person is ethically bankrupt to send arguably human entities into battle to be slaughtered, they're probably ethically bankrupt enough to just send the poor. Cheaper and more cost effective.
As for spies; well, it would be just as easy to engineer a rat to basically be remote control as it would be to give it human level intelligence. Plus, you wouldn't have to worry about the r.c. rat ratting you out if it got caught.
I haven't really gone into the question of human-to-animal; given tattoos and piercings, body modifications such as workable cat tails would probably be plausible. Whether they become common is more a question of fashion than science, though I don't think you could really call that "creating furries". More serious modifications, animal genetics as performance enhancement drugs of the future; well, that's probably the area to look into. However, once again, would these be really "furries?" Just because you're packing wolf genes does not mean you're packing a wolf snout. In fact, it would almost certainly be advantageous to conceal these advantages, just like it would good not to advertise your performance enhancing drugs today.
Like you say, the applicability is elsewhere. Oh, sure, people are going to do weird things to their bodies in the future, and weird things to animal bodies, but a true human/animal hybrid does not have a funcion. It would be an expensive, ethically wonky experiment that would serve no purpose, other than to say you did it (which is what I mean by doing it teaches us nothing).
I think it is kind of difficult to say something will be prohibitively expensive in the future short of something requiring massive amounts of energy due to laws of physics. No one will directly bridge the gap from current technology to such things, but people will develop more general tools for understanding body modification and genetics for other reasons. It would be difficult to say how far such things will advance, but at some point it could lower costs to well within the realm of "just for fun." Computer technology has advanced to the point that there is a large amount of R&D money spent on just things for entertainment, and tools are around for people to take advantage of that for the furry niche with video games. I think the harder thing to guess at is not how low such barriers can get, but whether some other technology will come up first that makes biological changes unnecessary for most people to get what they want out of that.
Good point. Could easily become cheaper in future.
I've already said why "expensive" is okay: the US military, and US government, take less flack for throwing away money than for throwing away lives. Engineered critters or critter/human hybrids would take the same role drones do now (just on land rather than in air).
A Predator surveillance drone costs $4-$10 million, depending on which source you read.
By comparison, the cost of raising a person (human or otherwise) to adulthood is about $0.3 million.
Regarding people being squeamish about sacrificing the lives of sapient nonhumans - there's ample precedent for using _humans_ as cannon fodder, for dangerous work, and so on. All that's required is for one group or another to feel that it has cultural or racial or religious superiority over the others. This happens quite a lot.
Expensive drones are not remotely human; furries are. You would not get ethical brownie points using furries. If you are unworried about ethical brownie points, why spend the money? Plenty of poor people.
Thank you for having a brain.
:/ Such a big video. 20mb for just over a minute.
"If all mankind minus one, were of one opinion, and only one person were of the contrary opinion, mankind would be no more justified in silencing that one person, than he, if he had the power, would be justified in silencing mankind."
~John Stuart Mill~
It was the HD version. I'll turn it off if that's a problem for people.
Is for me and there's not even the option for lower res versions.
"If all mankind minus one, were of one opinion, and only one person were of the contrary opinion, mankind would be no more justified in silencing that one person, than he, if he had the power, would be justified in silencing mankind."
~John Stuart Mill~
That's pretty incredible; particularly for professional costumers. Of course, it has a downside: where you once had to use wild gestures to express an emotion, you can be lazy and simply smile.
Of course, the moving eyes thing is pretty creepy. I always thought the "Follow-Me" eyes on fursuits were creepy enough, but this puts a whole new spin on it.
That and it's made in Japan, so with the stronger Yen it'll grow more expensive over here, unless they outsource it to other countries.
I'm kind of surprised this isn't already happening with multiple fursuiters. It wouldn't be too hard to do rather basic tracking of facial expressions with some of the tools already out there, although it would require some computer and electromechanical skills in addition to the normal efforts making a fursuit require.
The animatronic part is already being done by many fursuiters (has been for quite a while, and it's becoming more common). There's quite a lot of off-the-shelf animatronic hardware used by the film industry that's readily repurposed for this sort of thing (and quite a lot of hobbyist hardware, too).
The part that's new about this research is a system that reliably reads certain aspects of human expression. Speaking as someone who works in an image processing lab, that's a royal pain (there's a reason they don't have eyebrows working yet).
Fursuiters who have the expertise to build an expression-reading setup (software or hardware) and an interface to let that drive the animatronic controller, are the people who are rare. Usually they go with hidden switches or motion-slaving instead (movements of the arm, hand, or shoulder trigger an effect, or touching a button concealed in one's pocket).
It is a very nifty subject, and there are often interesting panels about it at cons.
Software for tracking certain basic parts of the face have been around for years now, including downloadable open source stuff and not just something hidden in academic realms. Since a lot of that is for the more general case of a fixed webcam with a person moving around, I would expect it to be much simpler if the camera could be put in a fixed position relative to the face as could be done with such a costume. Even adding some sort of tracking points like used in motion capture work wouldn't be too big of an effort compared to what some fursuits take to assemble. And it is not like such a system has to be perfect as long as long as there is some graceful degradation when uncertain as opposed to unrealistic expressions or changes.
i can't wait to see how this project progresses. :3 the other projects these researchers are working on are also impressive!
Rime, is your cat logo taken from the late-1980s "Fantadroms" sneezing-cat animated cartoons by Studio "Dauka" in Riga, Latvia? I didn't think that anyone else in America was familiar with them!
Fred Patten
why yes! i love older Russian animation, and Fantadroms is one of my favorites. thanks for noticing!
Actually, "Fantadroms" is Latvian animation, not Russian, although when these were made, before the fall of the Soviet Union, all animation business was centralized through the Soyuzmultfilm head office in Moscow. Each of the 15 Soviet republics theoretically had their own animation studios for home production, although I don't know how much animation actually got produced in Uzbekistan, Kyrgizstan, etc. Armenia did have a studio. Once the Soviet Union fell apart, the republics became independent and so did their animation studios, no longer supported by socialistic subsidies. I don't know how many survived under capitalism, but I think that Dauka in Riga did.
These areas are very nationalistic today. Latvians would probably object to Dauka's animation being called Russian.
I had just joined Streamline Pictures, which specialized in buying the American rights to Japanese anime, when someone in 1991 or 1992 offered to sell us all of Dauka's animation. That is how I came to see "Fantadroms", the "Ness and Nessie" theatrical feature about the Loch Ness monster and her mate, etc. Carl Macek, Streamline's boss, liked the animation but said that the offeror was connected with the Russian Mafia and did not have clear title to Dauka's films. The sale never went through.
Fred Patten
yeah i had forgotten that it was actually a Latvian animation, i probably should've said soviet-era animation instead of Russian. i'm by no means an expert on the subject but what i find on Youtube is really amazing! some of it is more like works of art than plain "cartoons".
thats a really interesting story too! that sounds like it would've been an amazing purchase, if it wasn't for the whole mafia thing. i'm sad to hear that it didn't work out, because Fantadroms is awesome. i hadn't heard of Ness and Nessie before, but i found it uploaded on youtube and i'm definitely going to check it out when i don't have to go to bed early.
i don't know when or if it will be done but i'd like to see this version of Fantadroms cleaned up. as you can see its really dark in a lot of places. :[ there IS a cleaned up version on youtube, but the music and sound effects are really different and a few of the scenes were omitted.
Post new comment