Some Eye-Popping Research From Siggraph 135
jamie found links to a discriminating selection of Siggraph papers at waxy.org. Among the more captivating: automatically improving the attractiveness of faces in portraits; automatic substitution of similar faces into photographs (with potential applications such as a privacy-enhanced Google Street View); and using still photographs to enhance video of a static scene.
Uncanny Valley (Score:1, Interesting)
Those altered faces are somewhere between Bunraku puppet and a prosthetic hand
Re: (Score:2)
Easy fix, after you run the face replace algorithm, just give it another round with the 'improve attractiveness of faces' one. Then you have your attractive replaced faces in no time.
Re: (Score:2)
Re: (Score:2)
Actually, I think the experiment only worked (with limited success) on male faces. And I am male. Something is wrong with the spacing of the female eyebrows and foreheads. Check the pics out.
I dunno -- I find the altered girls' faces more attractive, personally. Which probably implies that I'm closer to the mean than you are in my subjective tastes.
You can't please all the cooks all of the time. And the authors of this paper implicitly acknowledge that, as they use a statistical consensus to get an indication of what appeals to the majority of viewers. There's always going to be outliers.
FX Show Nip/Tuck said it best (Score:4, Insightful)
"Beauty is Symmetry, and you have none"
One of the main characters in the plastic surgery show Nip/Tuck made that comment. It seems as if TFA applies said comment.
Re: (Score:1, Insightful)
Re: (Score:2, Informative)
"Beauty is Symmetry, and you have none"
One of the main characters in the plastic surgery show Nip/Tuck made that comment. It seems as if TFA applies said comment.
I found the beautification piece interesting, and yes, symmetry and proportion are very important. Bigger girls that are still in the right wast-hip ratio (http://en.wikipedia.org/wiki/Waist-hip_ratio) are perceived as attractive, and the same goes for other body parts. The research in the article focuses on face symmetry, and some of the subtle
I sent them a pic of my face to be touched up (Score:5, Funny)
All I got back was an email that read "ROTFLMAO!"
Re:I sent them a pic of my face to be touched up (Score:5, Funny)
Re: (Score:1)
Maybe you are a denormal :-P
This one ? (Score:2)
http://media1.break.com/dnet/media/2008/2/14feb6-bitter-beer-face.jpg [break.com] :)
So in summary (Score:5, Insightful)
Re:So in summary (Score:5, Insightful)
that was exactly what I was going to write :) But the first guy sort of defeats that rule, he's actually a little broader in the face than the source image.
A simpler rule would be 'add symmetry', mirror the left half of the face (or the right half, flip a coin).
Adding a smile also goes a long way towards making people prettier, in fact a smile really is the best make-up.
Re:So in summary (Score:5, Interesting)
I can see this tool becoming very popular with the Myspace crowd once they realize the limitations of the current "hold camera above head level" method [officialda...source.com].
Re: (Score:2)
Well I'm not overweight and I don't think I'm ugly. But I have tried some shots with the camera up higher than me because I wanted to give the viewer (hopefully a man) the sensation of being taller than me and looking down at me. I figured guys like that. And I have a friend who did a high camera angle specifically to give a view of what's in her shirt. So there are at least two reasons for experimenting with high camera angles that have nothing to do with trying to deceive the person viewing the photo. And
Re: (Score:2)
NO! There are three billion of us, you can't say "guy's" like that. Is this just trite pedantry? Not really - why try to disguise what you are for the sake of one proportion when there's another proportion who will go for exactly what you are? I'm 6'1" (hopefully a man) and I go for girls who are tall (as a
Re: (Score:2, Informative)
Why is this moderated Troll?
Re:So in summary (Score:5, Interesting)
Re: (Score:2)
I think that plastic surgery is pretty horrible, I think if you think you need a plastic surgeon you really probably need to see a shrink or watch less TV (because that seems to be pretty much coded to give people low self esteem).
Sawing out your eye sockets and gluing them back in at slightly different angle causing all kinds of optical trouble does not seem to be reason enough to fix a minor perceived defect. What's so bad about diverity ? Celebrate the fact that you are uniquely you, and leave the plast
Re: (Score:2, Informative)
Careful inspection reveals that the woman's lips have been broadened, also.
Re: (Score:1, Funny)
My body does this automatically.
For instance, if I notice a pimple on the left part of my face, almost 90% of the time, I will have one on the same spot on the right!!
Re: (Score:2, Insightful)
The algorithm seems to be highly dependant on gender. For example women aren't allowed to have a broad chin, men aren't allowed to have a narrow chin. So what it does is enforce gender stereotypes, which is probably what the majority base their perception of beauty upon. Sad but true.
Re: (Score:2)
She works with the doctors who conduct trials for aesthetics (botox, restylane, etc) and in their conversations, a doctor had mentioned that while a MORE symmetrical face is desirable, a completely symmetrical face would not look attractive.
Maybe this would account for why some of these faces look a little off (esp the men
Re: (Score:2)
Interesting! So, apparently we are able to discriminate between 'perfect symmetry' (and label it for instance artificial) and nature's optimum (say > 95% 100% ?) ?
That's a pretty deep insight really, I wonder if there is embryological machinery that keeps us from becoming 'too symmetrical' ?
Re: (Score:2)
He didn't specify what a good amount of symmetry was, but intuitively, I'd guess it's closer to perfectly symmetrical than not. But what do I know - this guy is supposed to be among the best in the world.
He said they'd look "weird" with perfectly symmetrical faces. My friend is not an engineer, so I think she was satisfied with that and didn't ask 100 follow up questions as I would have!
It kind of makes sense, I guess. It's really not natural to have a "
Re: (Score:2)
I can smell a PHD in there somewhere :)
The title would run along the lines of 'optimum a-symmetry in the species of homo-sapiens to maximize chances during sexual selection'
Re: (Score:2)
Or, on a non-monetary side you'd make a great contribution to humanity and make great connections with doctors that always want their opinions to be heard (and probably mentioned in P
Re: (Score:1, Interesting)
Or even better, if it's a woman, make them naked. I've found this makes them so pretty you don't even notice their face.
So why is it that a naked face doesn't excite you? It's naked, after all.
Re: (Score:2)
So by your reckoning women should wear burqas, lest they be seen naked.
Re: (Score:2)
Well something like that, if you made it too thin it would also be ugly, there is some median symmetry, which if you deviate from (too large/too small) you get ugly. There is some statistical ideal each part of the face can have before it deviates too much.
Just like too fat vs too thin (i.e. starving kid in africa vs a fat person).
Re: (Score:1)
Just add symmetry and make thinner.
It will take more than that to improve the attractiveness of the average slashdotter.
Re: (Score:2, Insightful)
Not add symmetry.
Follow the golden rule.
Re: (Score:2)
Do you mean the golden ratio?
Although using your own willingness to have your photo modified could be an important prerequisite for doing it to others, to be polite.
dating sites will love this (Score:5, Interesting)
I wonder how soon they will be offering the "attractiveness improvement" service to the photos of their subscribers. I don't think they have enough CPU power to improve mine, though.
Re: (Score:3, Funny)
Why the hell would I need their services at all? I always just send a picture of Fabio like normal people do.
Re: (Score:3, Funny)
For the reverse process (to make other people look more attractive), I've developed my own tech. I call it Beneficial Ephemeral Eye Reticulation googles. Basically, a pair of B.E.E.R. Googles make even the attendees of a Linux kernel hacking conference look aesthetically acceptable. If only I could make the effect a lasting one...
Re: (Score:3, Interesting)
Who would want it? Surely, even butt-ugly people realize that eventually they'll have to meet their dates in person, and not being recognized is not going to be a desirable outcome. All those "more attractive" results look nothing like the original person. If I wanted to lie about my appearance, I'd sent a picture of Jean Claude Van Damme or something.
Re: (Score:2)
If it's a subtle tweak to Look Better it could be quite important. First impressions and all that. When they meet you in person, okay, maybe you don't look *quite* as good, but the chance that they'll realize it's due to digital manipulation is basically zero - and, honestly, getting the other party to meet you in person is half the battle.
Re: (Score:3, Interesting)
With this kind of technology you will never want
Re: (Score:2)
Reminds me of the earlier Sci-Fi series from the 1980's (Dr. Who, Buck Rogers in the 25th Century, Battlestar Galactica). They didn't have the prosthetics effects available that are available now, so they choose actors with distinctive features and costume design.
Re: (Score:2)
Strange 3D photo pairs (Score:3, Interesting)
I rotated the pairs of adjusted faces so they were left to right (and the faces were on their sides), and defocused my eyes as if I was looking at a 3D stereo pair of pictures to see what would happen. The slight differences made the portraits appear to me as if they had been photographed in 3D. The places that had been changed were subtly evident as a misalignment -- in the eyes of some, for example. I realize this is a fudged 3D effect, but might there be some use for it?
---
I write pointed political and business short stories at http://klurgsheld.wordpress.com/ [wordpress.com]
Re: (Score:3, Interesting)
Well, any two similar pictures can get that effect if you focus on them as though they are one. I occasionally use it to cheat on "Find the 10 differences" puzzles, but I don't really see much more use for it. :)
Re: (Score:2)
Back when I was having problems with epilepsy, my doctor sent me to an eye clinic to train my eyes to work together better. This training involved numerous exercises working with those magic 3D pictures and polarized glasses. There was one thing on the computer where you could see a bird with one eye on the screen, and a box with the other eye. You then had to get the box around the bird (It would chirp at you when you got it.)
Re: (Score:2)
I do this when I drink to try and restore some depth to the blur.
Ghost of Clarke seen skulking nearby. (Score:4, Insightful)
The first two are meh-worthy, but the last one approaches magic-grade technology. Wow!
Re:Ghost of Clarke seen skulking nearby. (Score:5, Insightful)
Re: (Score:2)
Thanks for posting that. I try to pay very close attention to Siggraph, I'm not sure how I missed that paper. Fascinating stuff.
Re: (Score:3, Insightful)
-- it's a marketable, easy to use technology that would be of huge benefit to typical consumers, yet chances are good it will never be commercialized.
I've noticed that a lot of SIGGRAPH papers will either only work for a small subset of inputs you would want to feed it or need a properly controlled environment to work and might need a lot of tweaking to get looking correctly. In my opinion, SIGGRAPH papers depend on demonstrating excellent best cases while mainstream consumer products require that the worst case be also acceptable.
Re: (Score:2)
That's exactly how it should be, in my opinion. Get the early results out early, and let the engineering of the finer details be done by as many teams as possible.
Re: (Score:2)
It's not that they get killed, so much, as the ideas just go out into the Wilderness of Beige to die because nobody seems to have the imagination to actually apply them.
They've spent an awfully long time thinking about how to productise Surface and Photosynth, for instance.
Re: (Score:2, Insightful)
Wow. No, really, wow! (Score:5, Interesting)
Re: (Score:2)
I'll never trust an image or video ever again. Never. Ever. Make sure you watch the "enhance video of a static scene" clip.
I suspect the main use of this technology will be myspace. Today, girls use strange photo angles to not like a 400 lbs swamp monster. With this technology, they won't have to.
I'm not sure if this is a good or bad thing.
Re: (Score:2)
You can say that again, the first two seemed like little more then an image warper combined with a facial recognition system. But that enhanced video one just about had me shitting myself with the things they managed to pull off.
Re:CAPTCHA (Score:2)
CAPTCHA is broken anyhow as there is no way to mitigate against the "porn site" attack.
Woz-inator (Score:1, Offtopic)
real footage? (Score:4, Insightful)
when cheap, easy, video editing allows this then supposedly real footage: news, family videos, wedding snaps will lose all veracity.
after every girl wants to look good for her wedding...
and before somebody says "it will never happen" this is only a logical extension of red-eye removal.
Re:real footage? (Score:5, Insightful)
I get the feeling from you that you trust it now. I find this confusing myself. Considering that an apparently large portion of Slashdotters very much consider themselves rationalist who do not believe things without proper evidence, it seems weird to me that many simply believe what they see in the news. These past week (maybe 2) there were at least two cases circulating around the internet where it had been observed that CNN has used footage from one event, trying to pass it off as that of another event. And that's pretty low tech.
News reports should be only be as trusted as logic can be applied to the report.
Take for instance the recent story of a Russian sipper shooting at a reporter. A few questions came to my mind:
News stories should be treated as untested pieces of evidence -- in most cases at least. The advancement of technology will only make it more difficult to tell truth from fiction.
Re: (Score:1, Informative)
What kind of sniper misses, and doesn't take a second shot?
The good kind (would you stay put if you just gave your position away?).
Aside from that, you make an excellent point about the trustworthiness of "news" "footage."
Ever seen "Wag The Dog?"
Re: (Score:2)
Not really. I saw a show about UK special forces snipers, and as part of their testing, they try to get off as many shots as possible from one location. First they just get in position and fire. If the (expert veteran sniper) trainers can't see them using binoculars from a sniper-range sort of position (200 meters or some such), another trainer moves toward the area and they fire again. If they're still not spotted, the trainer stands right b
Re: (Score:2)
Re:real footage? (Score:4, Informative)
What kind of sniper takes a shot and misses?
The kind that isn't a formal and TV depicted sniper. The kind that isn't a sniper at all.
Rather just some goon soldier or citizen with any old rifle from a fair distance away. Just becouse the media calls it a sniper, doesn't make it so.
Re: (Score:2)
What kind of sniper misses, and doesn't take a second shot?
Umm, the kind that had only one bullet? :)
Re:real footage? (Score:4, Insightful)
> What kind of sniper takes such a shot and misses?
No one is perfect. Long distance and wind variations can cause a miss.
> What kind of sniper misses, and doesn't take a second shot?
The smart sniper. There was no way a second shot would have hit. Everyone was moving around too much.
> How does one tell the affiliation of a sniper?
If they shoot at you, you can be sure it's the enemy. The sniper would have easily figured out which side the potential target was on.
> Do they sign their bullets or something?
Signing the bullet would have screwed up the ballistics. Snipers are extremely anal retentive when it comes to their rounds. They usually use hand loads and they buff the round to remove any imperfections.
FYI, a close friend was a sniper for SpecOps.
Let's hope there's a statute of limitations on it (Score:2)
If they shoot at you, you can be sure it's the enemy.
True, but as we saw recently with the anthrax researcher, your "enemy" doesn't necessary work for the other guys.
FYI, a close friend was a sniper for SpecOps.
And you're still alive? Whatever happened to "I'd tell you but then I'd have to kill you"?
Re: (Score:2)
Re: (Score:2)
"What kind of sniper misses, and doesn't take a second shot?"
The kind who uses a bolt-action rifle for accuracy and is careful not to give away his position by firing a second, less-likely-to-hit shot?
Re: (Score:1)
Are you kidding? I don't trust the [nytimes.com] news [theregister.co.uk] now!
With the right tech, in 20 years I'll be able to pick and choose whose face I want to deliver the "news", or star in my favorite shows/movies for that matter. Altered wedding photos will be the least of our worries. ;)
Re: (Score:3, Insightful)
You trust it now? Are you new here?
Re: (Score:1)
No it isn't. This tech makes the scene different from what it would look like to someone who was there. Red-eye removal makes the picture look like it would to someone there. Unless your eyes do actually glow red ...
Re: (Score:2)
Trust is foolish anyway. People lie all the time, unintentionally get things wrong, get deluded. The press are only human.
Re: (Score:2)
All the reasons that brought up the expression "dont believe everything you hear" are going to become relevant for video now.... any ideas for a new expression?
'dont believe everything.' is a little too depressing :P
Lists of Siggraph (and other) papers (Score:5, Informative)
I'd call this karmawhoring, but seeing as the editors didn't even bother linking to claimed list at 'waxy.org'... lists of Siggraph papers have been kept by Tim Rowley and Ke-Sen Huang for years. You can find this year's list at:
http://kesen.huang.googlepages.com/sig2008.html [googlepages.com]
And an overview of all years at:
http://kesen.huang.googlepages.com/ [googlepages.com]
This also includes lists of papers presented at other events such as Eurographics.
For even more fun, visit the papers' authors sites; they often also publish papers at seemingly unrelated events that contain some interesting computer graphics gems.
Re: (Score:2)
Thanks to your link I literally fried my graphics card :-(
I downloaded a program from one of the sites linked there, I think it was this [cmu.edu] but I am not willing to run it again on my replacement card to confirm
Re: (Score:1, Informative)
but i've not been able (ok, only 2 googlesworth;-) to decode this 3 letter acronym: NPR
NPR stands for Non-Photorealistic Rendering
whoa (Score:1)
One Possible Use (Score:3, Funny)
Re: (Score:2)
Actually, no. Or at least not yet... The skills etc involved in making the image manips look good are still far more expensive than the skills of porn performers. (well, most anyway)
Beauty (Score:2, Interesting)
The revolution will not be televised (Score:2)
In fact, it managed to get so different that got titled Clone Wars.Now you know why the president is so different in real life from what you see in TV.
wow (Score:2)
I'm still waiting to be able to buy a plenoptic camera [slashdot.org]
Interesting stuff... (Score:2)
Seems to me they could take this a step further and implement a sort of convolution matrix, but instead of modifying neighboring pixels, they'd be making subtle modifications to facial features.
Re: (Score:2)
Re: (Score:2)
Alright, my analogy wasn't appropriate. I was thinking of the Convolver function in a Painter plugin called Kai's Power Tools.
Pure genius (Score:2, Funny)
We took this spectacularly attractive blonde chick and wrote software that "automatically" turned her neutral facial expression (or slight scowl) into a smile. While we were at it we improved the symmetry of her face. Now she's gorgeous! We're brilliant!
But that's not all. We also took this African American guy and automatically plugged in a different African American guy's face that had the exact same pose. We're Gods I tell you!
Re: (Score:2)
It rather looks like an uglyfication. (Score:2, Insightful)
I could not decide, which versions looked better. I only recognized that they ere mostly non that good looking on both fotos.
Then I looked closer, because I know a bit about the methods behind it. And they did some big errors, like copying the one side of a face to the other, when the face did not look perfectly straight into the camera. This gave some weird results. Some faces even looked quite unnatural (especially, but not solely the focus on huge foreheads.
P.S.: I'm happy that I now since the last month
meh then wow (Score:1, Interesting)
The first two with the faces...none of them looked right after being touched up. That's probably because our brains are really really good at processing faces. They looked wrong, and creepy in some way to me. Not impressed with the results at all.
The static scene improvements with photo's though. Wow. Fooled my eyes everytime. It always looked better and when they started swapping out elements, hiding posts so you could see the flowers behind, fixing the shaky video was the only thing that looked at li
Easy. (Score:5, Informative)
Making faces more attractive is easy. All you have to do to get a reasonable increase is to make them more symmetrical.
If you want yet another increase, there is a set of ratios for distances between features that uncannily applies to pretty much everyone who is widely considered attractive. Shift everything closer to those ratios, and you'll get a big improvement.
Want more? Fix skin blemishes.
Between the three of those, you can make incredible strides. I would highly encourage any interested to watch "The Human Face".
It's not about making faces more attractive (Score:2)
Re: (Score:3, Interesting)
I'd like to see the face beautification software applied to fifty of the most beautiful actresses in Hollywood. I'd like to how much more beautiful they get, but also if some of them lose their unique look that makes them attractive to some.
Digital beer goggles. (Score:3, Funny)
Now she can look as good the morning after as she did the night before!
Found the female more "real" (Score:1)
does anyone realize the implications of this tech? (Score:3, Insightful)
Video does lie now.
Can video ever be trusted again where evidence is concerned?
Complicated Software (Score:5, Funny)
See the following example for how I was able to increase the attractiveness of an already attractive Hooters girl using only Microsoft Paint. (exported via Fireworks for filesize optimization)
http://img119.imageshack.us/img119/9474/hooters4si8.jpg [imageshack.us]
Re: (Score:2)
Thanks for the laugh.
Complicated Software (Score:4, Informative)
Moderators, thanks for the bigger laugh.
You can already get this (Portrait Professional) (Score:2)
Feed in a head-n-shoulders photo and it will automagically turn you into a supermodel
Payware but not very expensive
http://www.portraitprofessional.com/ [portraitprofessional.com]
More items from Siggraph... (Score:2, Informative)
A couple of items of interest displayed at Siggraph this year as well which I think have potential.
Microsoft's come up with a way of painting objects onto an object extracted from a video, then reinserted to the video that remains accurate when the camera angle's changed. Their research paper's called Unwrap Mosaics [microsoft.com], and you can see a video on Youtube here [youtube.com] (higher quality video on the reseach page).
A company called Image Metrics [imagemetrics.com] have made a video with actress Emily O'Brien [wikipedia.org], using Light Stage technology from
Looks Photoshopped (Score:2)
Yeah, the reflections are all wrong. Definitely photoshopped. [xkcd.org]