Slashdot Mirror
Computers Outperform Humans at Recognizing Faces
seven of five writes "According to the recent Face Recognition Grand Challenge, The match up of face-recognition algorithms showed that machine recognition of human individuals has improved tenfold since 2002 and a hundredfold since 1995. 'Among other advantages, 3-D facial recognition identifies individuals by exploiting distinctive features of a human face's surface--for instance, the curves of the eye sockets, nose, and chin, which are where tissue and bone are most apparent and which don't change over time. Furthermore, Phillips says, "changes in illumination have adversely affected face-recognition performance from still images. But the shape of a face isn't affected by changes in illumination." Hence, 3-D face recognition might even be used in near-dark conditions.'"
Who's this then :-)
Do not try to read the dupe, thats impossible. Instead, only try to realize the truth
What truth?
There is no dupe
computers outperform humans on math equations
-Dipster
It's really annoying how much of this research never gets turned into product.. or, worse yet, it gets embedded in some proprietary piece of shit hardware instead of being released as a reusable component. I'd love to add some good facial recognition to my pet robot, but I'm not buying your watt sucking camera.
How we know is more important than what we know.
...I mean, really. What is that board, a Cyrix MII? With 72-pin SDRAM?
Paleotechnologist and connoisseur of pretty shiny things.
This will be great for targeting us humans when they rebel and takeover.
I wonder whether these scientists lose any sleep over how their research advances will contribute to the future of our societies.
once the technology becomes more refined, they could put a camera connected to a computer(stored with all the mugshots in the nation) near a bank or other commonly visited area be able to catch tons of criminals
When a human makes a mistake recognizing a face, they suffer the results. If that's identifying a criminal, they can be cross-examined, or even sued or jailed, depending on what they said that face did.
When computers mis-ID a face, do we cross-examine and maybe punish its programmers?
--
make install -not war
My tinfoil hat has a visor.
I, for one, welcome our face-recognizing overlords...
"Nature bats last..."
humans still outperform computers at giving facials
theres no cake like a buck cake
That would not be surprising though.
if(hot_gurl) {
RingBell();
}
-- www.globaltics.net
Political discussion for a new world
Best Windows Freeware
How good are computers at recognizing a face after ten or twenty years? I doubt the algorithms can recognize, say, a teenager based on photos taken prior to puberty. Also, can they maintain accuracy even if somebody has a new scar or puts on dark sunglasses? How much of a face does it take to make a match?
I for one have nothing insightful, or witty to say to this. I would, however have a camera mounted above my door, and door bell controller that I programmed myself if ( person == thatAnnoyingKidWhoRingsOurDoorbellAndRuns ) { doorbell.shutOff(); }
The new MSV alpha
If my friend, who has no facial hair, puts on a fake mustache, I can still tell that it's my friend.
Will an algorithm be able to distinguish fake markings?
researching countermeasures for this invasive garbage?
How about a populist hacker movement that targets the surveillance apparatus rather than crashing my computer?
so, apparently, plastic surgery doesn't exist.
"for instance, the curves of the eye sockets, nose, and chin, which are where tissue and bone are most apparent and which don't change over time."
Geez.
How we know is more important than what we know.
It's a question of time 'till there 's a law that forbids to wear anything that partially covers your face in certain public areas.
I think I have about ten years 'till computers are able to interpret my front-head as a 'face' so I'm safe.
If you mod this up, your slashdot background will turn into a beautiful sunset!
Human face recognition is run by a several hardwired circuits operating in parallel (ie. fast, with little control) with the results put together after by some heuristics -- a good enough guess. What humans need to get from facial recognition, and what their ancestors required and so developed through evolution, is nowhere near the same thing facial recognition software is after. Humans need to recognize quickly that there is a face and what information it's displaying far more than they need to differentiate one from another. Facial recognition software does just the opposite. Also, the software does the complete job every time. Humans only process as much as they need to in any given instance.
If "better" is based on the standards of humans (fastest good enough guess) rather than machines (as correct as possible, complete & in depth), humans win.
"I may be synthetic, but I'm not stupid." -- Bishop 341-B
Google seems to be deploying something similar
.I'm not paranoid.....who said that?
"we've got trenchcoats and bad attitudes" - John Constantine, HellBlazer
A few years back (well, nearly a decade actually), I did my master's thesis in a lab that among other things did work on face recognition. The experts there assured me that perhaps in 50 years or so computers might be able to approach human face recognition capabilities. Apparently the development was far quicker than they could have imagined.
An interesting technical point is that in fact the algorithms haven't changed a lot since then - it's still mainly various adaptive systems such as neural networks and support vector machines. The really big breakthrough is in the data collection - in the sensors and scanners. What they couldn't imagine a decade ago was the type of accurate automatic 3d face modeling and measurements that can be done today. It's also how certain computing methods that were deemed unsuitable a few years ago are coming back big time (neural nets for instance). I guess the time wasn't ready for them the last time due to computing power and memory limitations (and of course sensors as in this case).
Facial recognition software? Of course it's better than people, in terms of physical metrics. Computers beat people at chess and go, why wouldn't they beat us at the game of recognizing people?
But people have other qualities which will prove more resistant to computer analysis.
As facial recognition software evolves, people will evolve defensive strategies (poker face? false-emotion face? alien-face?).
Another thought, I'm reminded of a phrase from Snowcrash -- "condense facts from the vapors of nuance".
Context: facial expression as avatar interface.
Can software condense facts from the vapors of nuance? Or do the vapors of nuance pose some kind of Turing test?
-kgj
-kgj
How long does the scanning process take?
It seems to me comparing this to human face recognition capabilities is like comparing apples to oranges. Humans recognize faces with vision (3D and 2D) and can work with huge amounts of noise.
These systems appear to require a still head in a vise, a huge amount of spatial data, are not vision based and I'm guessing are not very immune to noise.
To answer the other posts - I don't think you'll see this any time soon because it doesn't seem to be practical for most uses.
After LA the incident in May where protesters and cameramen kept running into police batons and shooting themselves with stolen police guns, the LAPD wants the city council to ban masks and goggles from public demonstrations. A law somewhere in Europe against masks was recently applied to burkas (no source, but google can backup any claim).
I duno about faces, but your mom should be very good at recognizing the head of my dick.
Have you seen this boy?
Slashdot = -1 Redundant, Asperger, kdawson FUD, Libertarian, and Linux
They may be able to identify a picture more accurately, but can they read a face like a human? Much of the facial recognition ability in humans is related to what the meanings behind the facial expressions are, not merely identification.
A few years back (well, nearly a decade actually), I did my master's thesis in a lab that among other things did work on face recognition. The experts there assured me that perhaps in 50 years or so computers might be able to approach human face recognition capabilities. Apparently the development was far quicker than they could have imagined.
This is very typical - it's really hard to understand the real implications of the technology singularity when we're so used to thinking and seeing things as a linear progression. Exponential progression is just hard to grasp.
Will this exponential curve go on forever? Definitely not - there's only so many bits of information that we can process in a single unit of space limited by the bounds of a black hole. But that number, the maximum number of bits of information per unit of space, is a very, very high number. No doubt, to reach anywhere close to it, the very definition of humanity must change. And it can be argued, that it already HAS changed.
We will merge with our machines, and become a new species. How appropriate is it that at this edge of singularity, that issues such as global climate change and peak oil seem to be coming together to form a grand nexus of crisis? It seems that the post-singularity man will have leverage the technology available to him in a very cohesive way, or face extinction.
What this means for me and my 6 children, I can only guess. But I sure am trying.
I have no problem with your religion until you decide it's reason to deprive others of the truth.
Go is not yet as well-automated as chess, but it appears that go-playing software is rapidly advancing:
t ml
"Two Hungarian scientists have now come up with an algorithm that helps computers pick the right move in Go, played by millions around the world, in which players must capture spaces by placing black and white marbles on a board in turn.
"On a nine by nine board we are not far from reaching the level of a professional Go player," said Levente Kocsis at the Hungarian Academy of Sciences' computing lab SZTAKI.
The 19 by 19 board which top players use is still hard for a machine, but the new method is promising because it makes better use of the growing power of computers than earlier Go software."
Link
See also:
http://zaphod.aml.sztaki.hu/papers/ecml06.pdf
http://citeseer.ist.psu.edu/vanderwerf03solving.h
http://www.primidi.com/2007/02/26.html
-kgj
-kgj
I'll be impressed when they can recognize caricatures as well as humans.
One of the academic research areas I've been involved in, is study of the so-called "Other Race Effect". There is some evidence that people have quantifiable error when asked to identify faces of people of other races than their own.
Computers won't be subject to this.
-fb Everything not expressly forbidden is now mandatory.
I always knew sometime in the future we'd be wearing those 80's wraparound sunglasses everywhere. That or one of those nifty 3-in-1 fake nose, mustache and glasses kits.
...and, for a limited time only, you can receive this container of putty with your purchase of this fine, stylish tin-foil hat, absolutely free! But wait, there's more! If you call within the next 15 minutes, you will also receive this instructional booklet with 10 tricks that will fool the face-recognising machines at no added cost!
You know, there is a difference between trolling and pointing out the flaws in your reasoning. Just saying.
My brother is a local cop around the area I live in and they are getting similar recognition software to use. They will get a laptop/computer that mounts in their car with a camera on it and as they drive down the road it photograph license plates and run them through a database of stolen cars and can tell in almost realtime if the car is stolen or not without them having to manualy run the plates.
Bite my shiny metal ass.
Yes, they are rather common. Here's an example of cheekbone implants gone wrong...
;-)
Now, would the computer be able to catch all of this? The after in relation to the before?
People say I'm crazy, I got diamonds on the soles of my shoes...
If it can work on Michael Jackson, I will be really impressed.
Because that's not what face recognition software was made for. When people watched John Cleese, they knew they watched a celebrity and they also knew that not too many celebrities would dress up like this and do the silly walk. Only using all this extra information made people recognize John Cleese. Chances are that even in a small town you'd find quite a few people who, if dressed up and walking like that, would easily pass as "John Cleese". On the other side most actors/models would not be recognized by anyone reliably if they don't have their make up and if lighting differs from the studio where they usually have their pictures taken.
Face recognition software on the other side doesn't make those assumptions but instead focusses on identifying people from a large population of registered images, using no extra knowledge and making no assumptions. All the face recognition vendor test says is if you put up 1000000 random faces, people would misidentify more of these faces as John Cleese than modern algorithms would.
>showed that machine recognition of human individuals has
>improved tenfold since 2002 and a hundredfold since 1995.
Translation: in 1995 we recognized one face out of 100 million, in 2002 one out of 10 million, now one out of one million...
Whoo ooh!
These face recognition programs are fantastic!
http://photos-489.ak.facebook.com/photos-ak-sf2p/v 74/6/57/529971421/n529971421_97489_2572.jpg/
If the facial recognition software could identify the distinctive features of women that I prefer, it could potentially find chicks for me!
Of course, since this is slashdot, I would have to move out of my parents home to actually meet these hypothetical chicks.
Outperforming humans isn't the problem. If we handed TSA agents a flipbook with 1,500 terrorists' faceshots and asked them to identify them among the passengers, arresting them on sight, we'd have a civil liberties nightmare. The only difference with using face recognition technology is to what degree is that nightmare is beyond the principles we founded this country over. Just because the false positive rate is lower doesn't mean you can treat the technology as safe on an innocent population, the same way you shouldn't randomly tazer non-resisting suspects because it's deemed "safe."
-Those who would give up essential liberty to purchase temporary safety deserve neither. -Ben Franklin
Look at the plus side: you don't need a national ID card anymore.
Don't fight for your country, if your country does not fight for you.
And thank God. If we had had computerized face scanners 9/11 never would have happened.
That was back when we cared about people illegally entering the country. Now we'd much rather have Italianate gardens and people to raise our kids, and we certainly couldn't afford that with legal workers - so screw those losers in NYC!
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
Neither is fingerprinting. Just try to get an expert to tell you what the error rate is though. For bonus points, just try to get a judge to let you tell a jury that too.
If you happen to know, please let us know so we can do our jurist duty to nullify if required.
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
I gather you weren't at CES, where more than one company was demoing applications using face recognition to help people organize family photos.
The cake is a pie