Slashdot Mirror


The Face Detector

Roland Piquepaille writes "Almost all human faces have common characteristics, such as two eyes and one mouth. Still, some people, affected by face blindness, cannot recognize one face from another one. So it's understandable that face recognition is a major challenge for computer vision systems. In "Facing facts in computer recognition,", the Pittsburgh Post-Gazette reports that a team from Carnegie Mellon University's Robotics Institute has developed a very accurate software to find faces within images. By analyzing only 768 pixels, the system can detect 93 percent of the faces in a set of images while falsely identifying four objects as faces. The Face Detector Demo is available online and you can submit an image for analysis and receive the results by e-mail. The technology will be used for security purposes, but also by digital photography companies who want to automatically reduce "red eye" effects. You'll find more details and references in this overview."

241 comments

  1. Maybe they should upload it... by Black+Parrot · · Score: 4, Funny

    ...to the Mars rovers.

    --
    Sheesh, evil *and* a jerk. -- Jade
    1. Re:Maybe they should upload it... by Doesn't_Comment_Code · · Score: 1

      Or test it out on Tammy Faye Baker...

      Which result would be correct?

      --

      Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    2. Re:Maybe they should upload it... by Anonymous Coward · · Score: 0

      In Soviet Russia,
      rover uploads you!

    3. Re:Maybe they should upload it... by Anonymous Coward · · Score: 0

      Either way, alien dirt substance detected.

    4. Re:Maybe they should upload it... by etheriel · · Score: 1

      apparently that wouldn't be of much use.

  2. Are you sure?? by pair-a-noyd · · Score: 1, Funny

    "Almost all human faces have common characteristics, such as two eyes and one mouth"

    Unless you live near Love Canal or TMI.....

    1. Re:Are you sure?? by Frymaster · · Score: 1
      Unless you live near Love Canal or TMI.....

      or unless you wear some particularly large sunglasses.

      you recognize this man as elton john. but would a computer be able to identify him with those glasses on? hm.

    2. Re:Are you sure?? by Anonymous Coward · · Score: 0
      HEY! I've lived near TMI all my life and I can assure you I have two eyes and one mouth (a big one according to some people). I was outside every day delivering papers while the fiasco was ongoing. During the winter months, from my bedroom window, I could see the tops of the cooling towers over the hill because the leaves were off the trees.

      Just because people lived near TMI during 1979 does not mean they're freaks. Though I must admit not having to use flashlights at night is a good thing.

      For the humor impaired, I am aware the original poster was being funny. The information posted above is true with one minor exception.

    3. Re:Are you sure?? by Anonymous Coward · · Score: 0

      Well tomorrow, we will see if it does! I just uploaded the picture of Elton John :-)

  3. But does it detect... by lukewarmfusion · · Score: 0, Redundant

    ...mountains on Mars as faces?

    "Sir, we've got a convicted felon... wait, that's just a bunch of flowers. I hate this stupid program."

    1. Re:But does it detect... by pridkett · · Score: 4, Informative

      Yes, yes it does. This one of the big problems with the software, is that some things look like faces and really aren't. A human can tell because we've got a lot more training on different data sets. After seeing some of the demos of this stuff, either they really jacked up the accuracy in the last few weeks, or it was under more controlled settings. Off a picture from a new york street it could only pick up about 60% of the faces and had a decent amount of false positives.

      Also, for those who won't read the article, this is just about finding the faces, not recognizing them. This is a prerequisite toward ubiquitous facial recognition.

      --
      My Slashdot account is old enough to drink...
    2. Re:But does it detect... by NanoGator · · Score: 4, Insightful

      "This is a prerequisite toward ubiquitous facial recognition. "

      Also good for video compression. Imagine if the codecs detected where the faces in a video were and intentionally avoided compressing them to hell and gone.

      --
      "Derp de derp."
    3. Re:But does it detect... by Anonymous Coward · · Score: 0

      Are you sure you don't mean "mountains on Mars as feces?"

    4. Re:But does it detect... by Short+Circuit · · Score: 2, Interesting

      Or stored a master and rebuilt the original on height/width/orientation.

    5. Re:But does it detect... by Anonymous Coward · · Score: 0
      Off a picture from a new york street it could only pick up about 60% of the faces and had a decent amount of false positives.

      This is New York spesific problem. So far only human faces are recognized.

    6. Re:But does it detect... by Anonymous Coward · · Score: 0

      Actually, no it didn't. Maybe it would on a cleaned-up version of the picture (the site is too hard navigate already, nevermind with the slashdot effect), but, like you said, it should be detecting it.

    7. Re:But does it detect... by bhtooefr · · Score: 1

      Well, http://vasc.ri.cmu.edu/demos/faceindex/05062004/us ers/2248.html (scroll to the bottom left) did something perfectly - look at Bush - according to the computer, he's two-faced. I could have told you that, but at least the computers can figure it out. I think they're going to hook me up to the Matrix...

    8. Re:But does it detect... by SuiteSisterMary · · Score: 1

      Humans have the advantage over computers, I think, because we're capable of seeing qualities, not properties.

      A human, for example, can grasp the essential 'chairness' of a given object, whereas a computer can look for a seat, four legs, and a back. We, however, can see a chair in a rock, the floor, and so on.

      Similarly, we can see a face in the clouds, a rock on Mars, a hint of shadow. A computer, on the other hand, can look for two eyes, a nose, and a mouth.

      We also have the uncanny ability to 'fill in the blanks.' Analog (i.e. values other than 'yes' or 'no') processing coupled with intuitive leaps; I don't know if it would be possible to simulate these.

      --
      Vintage computer games and RPG books available. Email me if you're interested.
    9. Re:But does it detect... by timjdot · · Score: 1

      Nice. I remember when the Face-It software won best in show in Comdex in like spring of 1994.
      I think the name was "Face-It".

      --
      Expect Freedom.
    10. Re:But does it detect... by Hentai · · Score: 1

      How... Aristotlean.

      Yes, it's absolutely possible to simulate these. In most current cases, we just don't have sufficient processor power to do so in real-time yet.

      Go read Douglas Hopfstaedter's writings sometime, it's fascinating stuff.

      --
      -Hentai [in vita non pacem est]
    11. Re:But does it detect... by SuiteSisterMary · · Score: 1

      Thanks for the reference; I'll add it to my Mighty List of Things To Read.

      --
      Vintage computer games and RPG books available. Email me if you're interested.
  4. Portable face detector by nizo · · Score: 4, Funny

    What I really want is one that is portable that will whisper the name of people into my ear so I never have to remember anyone's name ever again. Something with hooks into the FBI's most wanted list would be nice too (Hey you just walked by a guy who is worth 2 million if you turn him in).

    1. Re:Portable face detector by Cruciform · · Score: 5, Funny

      Man, that would rock!

      Though it might be a bit disconcerting to the people who observe you screaming "JACKPOT!" and jumping up and down on a seemingly innocent pedestrian.

    2. Re:Portable face detector by ad0gg · · Score: 3, Funny

      Or it could hook into the "Girls who put out on the first date" list so I know who to hit on at the bar.

      --

      Have you ever been to a turkish prison?

    3. Re:Portable face detector by Anonymous Coward · · Score: 0

      Excellent !
      Have others ?

    4. Re:Portable face detector by Xentax · · Score: 4, Informative

      I think this technology just recognizes faces from backgrounds, it *does not* appear to uniquely identify faces (a la fingerprints).

      Others have tried that, and we all know how monumentally insufficient it has been thus far as a legitimate security tool, in terms of missed matches and a high false-positive to actual positive ratio.

      Xentax

      --
      You shouldn't verb words.
    5. Re:Portable face detector by D-Cypell · · Score: 1

      Good idea! ...and if they are looking for somebody to research the data for this application, I would like to be the first to volunteer!

    6. Re:Portable face detector by BLAG-blast · · Score: 4, Funny
      Though it might be a bit disconcerting to the people who observe you screaming "JACKPOT!" and jumping up and down on a seemingly innocent pedestrian.

      The false positive rate be a little anonying...

      --
      M0571y H@rml355.
    7. Re:Portable face detector by Lord+Agni · · Score: 5, Insightful

      Bruce Schneier points out the problem of false positives in his "Secrets and Lies", and you'll get the same argument in any freshman statistics class: If the target population (of identified terrorists in the country, or people with AIDS) is extremely small, the probability of a false positive is greater than the probability of a true positive. If this system is correct 93% of the time, it's wrong 7% of the time. How many terrorists are there in the average metropolitan airport? I'd say zero (on average) How many people would be incorrectly identified? 7%. An airport with, say, 2000 people in it would have 140 of them misidentified. Even if the average airport had 20 terrorists, the false positives or misidentified would greatly outnumber them. And since the system is wrong 7% of the time, one or 2 of those 20 terrorists would be misidentified as not being a terrorist.

    8. Re:Portable face detector by Martin+Blank · · Score: 1

      Look at it from a stepped approach:

      1. Face detection software identifies faces from background.
      2. Face identification software examines multiple aspects, retrieves matching information from database.
      3. System plays back critical information into a hidden earpiece.
      4. (Optional for those with glasses or ocular implants) System displays known face in field of view for confirmation. An advanced system could do the same thing with multiple faces in a background, adding a name over each person as they're identified, and then providing more detailed information as a person comes closer.

      Such a system would be useful in business, social situations, and with individual security personnel. A guard or cop could have this kind of system in place for cross-checking on the ground. An initial scanning system could even locate people and dispatch personnel for final checking up-close before any actions are taken.

      --
      You can never go home again... but I guess you can shop there.
    9. Re:Portable face detector by Dasaan · · Score: 1

      Hey you forgot

      5. ...
      and
      6. Profit!

      --
      XP is basicly 98 with a lot more extra features to hunt down and disable. --Dram
    10. Re:Portable face detector by Martin+Blank · · Score: 2, Funny

      I thought about it, and figured someone would say something about it, but decided I wanted to keep my post serious. I'll compensate by pointing out how a proper implementation would require a Beowulf cluster of Debian boxes so as to be able to pick out CowboyNeal in the crowd of Natalie Portman clones from the background of hot grits.

      Happy? :)

      --
      You can never go home again... but I guess you can shop there.
    11. Re:Portable face detector by SmackCrackandPot · · Score: 1

      Maybe you could have an optical implant like Geordie in ST:TNG . If the ships's computer can tell where everyone is located, and his eye-visor can provide additional information beyond normal human vision, it wouldn't be too difficult to have the name, rank and status of all people within visual range visible. Maybe as semi-transparent text above or slowly rotating around that person.

    12. Re:Portable face detector by Cruciform · · Score: 1

      Yeah that's frustrating, kind of like scraping the silvery coating off a lottery ticket to find "Good luck, try again." except this lottery ticket screams.

    13. Re:Portable face detector by the_mad_poster · · Score: 1

      Yea, but after jumping up and down on a couple of innocent pedestrians, YOU'LL be one of the innocent pedestrians that gets jumped up and down on.

      The false positive rate would eventually even itself out nicely over a larger population as more and more people made the "Most Wanted" list for attacking innocents and left fewer and fewer people to catch them and turn them in.

      --
      Alito: A vote for Alito is a punch in the eye to put that bitch back in her place!
    14. Re:Portable face detector by DJStealth · · Score: 1

      I don't think this is anything new. I have an IEEE paper here of a group that did a similar system in 2001. There are systems that do face recognition and verification; but nothing with high accuracy (people have been working on this for years using many methods, the main problem is that you need to be facing the camera perfectly straight). Iris and retina recognition, on the other hand, can be done with a relatively high accuracy.

    15. Re:Portable face detector by MurphyZero · · Score: 1

      You are absolutely correct about the effects of a large false positive rate. However, remember that the false negative rate is not necessarily the same as the false positive rate. In fact, such a system as you described could be 100% accurate in recognizing all 20 terrorists. Now you would have identified 160 terrorists, but only 20 are actually terrorists. And Ashcroft would keep all 160 in jail indefinitely. As you stated, with a nonzero false negative rate you may have terrorists escape detection.

      Offtopic, this effect also concerns some innoculations (not all). The vaccine may prevent the disease in 99% of those who would have gotten the disease. However, it may cause the disease in 0.001 percent of those who take the vaccine, who would not have gotten the disease otherwise. In a population of 100 million, with an infection rate of 0.1%, 100,000 would normally get sick and only 1000 will after the vaccine. Of the 99.9 million who not have gotten sick, 999 will get sick because of taking the vaccine. So, while the vaccine saves 99000, half of those that get the disease would get it due to the vaccine. And this is with a vaccine that would be consider 99.999% safe.

      Now if instead of a vaccine, it is a terrorist detector, given to everyone. Even with a very accurate detector, detection with have to be treated very carefully.

      --
      Our founding fathers removed the guys in charge. Be American. Vote incumbents out.
    16. Re:Portable face detector by walt-sjc · · Score: 1

      I just need one that tels me which waitress is mine... Is it just me or do they all look alike???

    17. Re:Portable face detector by Arkaein · · Score: 1

      Not to mention that terrorists are far more likely to take measures to alter their appearance, and the system can only identify known terrorists/criminals.

      Even if the systems work within acceptable error range without considering these factors, once these are accounted for the recognition systems really look pointless.

    18. Re:Portable face detector by Anonymous Coward · · Score: 0

      Yes, but like defective voting machines, some goose has already bought obsolete face identifying software for airports that can't possibly do the job. MIT stuff rocks, so if the airport stuff has >10% false positives, no wonder the airline industry is in a spin. Hassles and longer queues = less bums on seats.
      The processing needed to get to 0.2% should be whats being talked about.

    19. Re:Portable face detector by subtropolis · · Score: 1

      Yes, correct. Though i think you missed the point of grandparent's first sentence - the software is, in fact, for face detection.

      I don't what the state of the art is for face recognition is*, but i suspect that the point of the originally mentioned software is as more of front-end to the recognition progs. A 7% false-positive rate for the detection simply means a handing an image of a lamp (or dog, or someone's arse) to the recognition routines. Only then will it compare it to those of known 'terrorists'.

      Anyone falsely detained for that is another problem with the system altogether.

      * For various reasons, i'm not so very optimistic about the whole thing.

      --
      "Our interests are to see if we can't scale it up to something more exciting," he said.
    20. Re:Portable face detector by whereiswaldo · · Score: 1


      Haha, that's awesome. You must be into stats. :)

    21. Re:Portable face detector by Cruciform · · Score: 1

      It would be just like the 'Life' simulator. What pretty patterns it would make as society crumbled :)

    22. Re:Portable face detector by Dasaan · · Score: 1

      Glad I didn't disappoint you.

      ok, ok I'm going home now

      --
      XP is basicly 98 with a lot more extra features to hunt down and disable. --Dram
    23. Re:Portable face detector by JackCroww · · Score: 0

      Or a monocle: http://www.mvis.com/

      --
      "Ayn Rand is a bloody socialist compared to me." - Robert A. Heinlein
  5. I know... by lcde · · Score: 5, Funny

    Where's Waldo? :D

    --
    :%s/teh/the/g
  6. a real use for this kind of technology by hype7 · · Score: 5, Interesting

    instead of the endless "let's use it in airports" crap, it looks like they've found a real use for this kind of thing.

    give it to blind people so they can know who they're talking to. But don't stop there - man, the number of times I've forgotten names... it'd be great if they could integrate this kind of thing into some glasses, that popped up the name of the person as you looked at them (assuming, of course, you knew them).

    whoever commercialises that tech first is going to make a lot of $$$, I think...

    -- james

    1. Re:a real use for this kind of technology by croddy · · Score: 1

      so, I understand you suffer from "voice deafness"?

    2. Re:a real use for this kind of technology by Anonymous Coward · · Score: 0

      nah, just couldn't type faster than the guy who posted above me ;)

      -- james

    3. Re:a real use for this kind of technology by Dark+Lord+Seth · · Score: 3, Funny

      Ah, something like glasses with a HUD in them?

      *beep* Jonathan McPherson. Age: 34 Occupation: Busdriver.
      *beep* Ellen Craigh. Age: 27 Occupation: Systems Developer.
      *beep* Unknown woman. Age: Unknown Occupation: Who cares? I'd hit it!

      I sense business potential if one can add random data as well!

    4. Re:a real use for this kind of technology by Doesn't_Comment_Code · · Score: 5, Funny

      give it to blind people so they can know who they're talking to

      Two guys walk into a convenient store.

      Angry man 1: "Empty the cash register into this back now mutha f'er!"

      Angry man 2: "Get down! _shots-fired-into-air_ All of you Get down on the floor!"

      Face recognition computer to blind man: "You are now speaking to Ronald Reagan and Ronald Reagan."

      --

      Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    5. Re:a real use for this kind of technology by NanoGator · · Score: 2, Insightful

      "instead of the endless "let's use it in airports" crap,"

      The "use it in airports 'crap'" happens because they're trying to get gov't funding to continue development. It's not crap to try to get ahead in this world by fulfilling a need. Don't like it? Point at Uncle Sam instead of the company trying to earn money.

      *Note: I'm talking in general, not specifically about this company.

      --
      "Derp de derp."
    6. Re:a real use for this kind of technology by mirko · · Score: 1

      Only if a Zilog Z80 assembly language listing along with a target appear on his/her face :)

      --
      Trolling using another account since 2005.
    7. Re:a real use for this kind of technology by drudd · · Score: 1

      I remember a Scientific American where they were at an MIT AI lab where they had something pretty close.

      Everyone wears a little pin radio transmitter, then when you are looking at a person, a special set of glasses you wear pops up their name in the lower right corner. Not enough to really see, but enough to help spark name recognition.

      So they got out of the face recognition problem, which seems harder than it's really worth.

      Doug

      --
      Venn ist das nurnstuck git und Slotermeyer? Ya! Beigerhund das oder die Flipperwaldt gersput!
    8. Re:a real use for this kind of technology by thefirelane · · Score: 2, Insightful

      give it to blind people so they can know who they're talking to

      Have you never had someone say "Hi" after calling you on the phone.... and known who it is?

    9. Re:a real use for this kind of technology by Anonymous Coward · · Score: 0

      Ok, I'll bite. I don't get it. Seriously.

    10. Re:a real use for this kind of technology by egomaniac · · Score: 1

      Ok, I'll bite. I don't get it. Seriously.

      The implication is that the robbers are wearing masks of Ronald Reagan. Watch the movie Point Break, it will make more sense.

      --
      ZFS: because love is never having to say fsck
    11. Re:a real use for this kind of technology by jhines0042 · · Score: 1

      Sorry.... Blind people know who they are talking to, because generally they don't need to know them by their face, just by their voice. Since they cannot see (and many have never seen ever) they develop a sense of what is going on around them with their other senses.

      Now when the blind can listen to a speach imparied person by way of sign language recognition, that would be cool.

      Presumably the blind person could talk to a hearing imparied person by way of voice recognition software printing out what is said.

      --
      42 - So long and thanks for all the fish.
    12. Re:a real use for this kind of technology by Anonymous Coward · · Score: 0

      man, you're stupid. go kill yourself k?

    13. Re:a real use for this kind of technology by Jardine · · Score: 1

      give it to blind people so they can know who they're talking to

      I suspect blind people use the same method I do for telling people apart, the sound of a person's voice.

    14. Re:a real use for this kind of technology by Threni · · Score: 1

      > real use for this kind of thing.

      Hmm. In parts of the UK they're testing out cameras which can read number plates and alert the police if they are stolen or belong to people they're after. Similar technology exists in some police cars. There are so many people they're after that they let loads of known villains get away because they're after bigger ones!

      Give it a few years and these cameras will be *all* over the UK, then the rest of the world.

      Give it a few more years and the same will exist for people - first at airports and train stations, but eventually the technology will be available as an add-on for any cctv camera/footage. You won't be able to go anywhere without people knowing where.

  7. Old news by Geoffd1 · · Score: 3, Interesting

    This is old news - software that finds faces has been available for years. To cite an older example, the company Miros, which later became TrueFace - they used a neural-net approach.

    1. Re:Old news by ViolentGreen · · Score: 1

      This is old news - software that finds faces has been available for years.

      Yeah, but face-recognition has never worked very well. Whether this does or not is yet to be seen.

      --
      Not everything is analogous to cars. Car analogies rarely work.
    2. Re:Old news by mobby_6kl · · Score: 1

      Yes, even this particular project dates back to 2001 (mabye earlier, I didn't RTFA) - scroll down their gallery page.

    3. Re:Old news by koninc · · Score: 1

      Its also some of the basic underlying technologies for computer vision-based eye and head tracking. This stuff needs to auto-acquire, then from that, find the eye (which is a lot trickier - especially to do accurately) and then track that. Computer vision company Seeing Machines is doing this stuff right now.

  8. Interesting indeed~ by Leffe · · Score: 0

    Let's see how well it is able to handle this image :)

    The results will be very interesing, does anyone know how long it is until 'night'?

    1. Re:Interesting indeed~ by bhtooefr · · Score: 1

      Quick question before I open the tab - is it work safe?

    2. Re:Interesting indeed~ by Leffe · · Score: 1

      It sure is. It's a colored manga artwork from some unknown series.

      Anything at 4chan in /a is safe. /c is safe too, /cm quite safe, ... yeah, I guess that's about it :)

    3. Re:Interesting indeed~ by bhtooefr · · Score: 1

      I opened the tab and got a leech message, even after turning off referrer logging.

    4. Re:Interesting indeed~ by Leffe · · Score: 1

      Argh, they must've implemented that yesterday :/

  9. Face detector by sotonboy · · Score: 4, Interesting

    I dont get the point ? It doesnt recognise faces, just tells you if theres one there. Thats not exactly state of the art is it ? When other companies are producing systems that can identify people from images, albeit inaccurately.

    1. Re:Face detector by SquadBoy · · Score: 1

      and therein lies the rub how do we know that if we give some of those other systems a picture of bacon and eggs arranged in some vague face type config that it won't tell us that it is Osama. Keep in mind the only data we have on most of those systems are what the people trying to sell them are telling us. Me I have my doubts that they have ever been tested to see if they can even identify a face.

      --

      Cypherpunks: Civil Liberty Through Complex Mathematics. Those who live by the sword die by the arrow.
    2. Re:Face detector by Anonymous Coward · · Score: 1, Interesting

      You are right to notice that facial recognition != facial detection. The latter is a much harder problem, but the Slashdot introduction for this article implies they are one in the same. What they've done is like talking about a computer program that can bring about world peace, then link to an article on Eliza (see: http://www.manifestation.com/neurotoys/eliza.php3) . If anything, this research just shows how difficult the problem of face recognition is. The research endeavor that is facial recognition is a blackhole for funding. Smart security analysts think that the research money would be better spent on paying more human analysts to do intelligence work. Here is what technology security pundit Bruce Schneier wrote on the subject after September 11th: http://www.schneier.com/crypto-gram-0109a.html#3.

    3. Re:Face detector by It'sYerMam · · Score: 1
      However, when saying that, you should take into account the following:

      How many times have you turned around and been SURE that you just saw someone you know, only to call to them, have them turn round and ask, "who the f*** are you?!"
      So you see, humans aren't all that foolproof at face matching. The only reason they're half-decent is because they're brought up with faces (neural net approach, anyone?) hence why it asian/japanese faces can often all look basically the same to a westerner.

      Along with that - think of a cartoon character. Seems likely that you won't be thinking of someone with a photorealistic face, and yet you'll probably be able to recognise it as a face. In fact, anything 3 splodges arranged in a triangle can be interpreted as a face by the human brain.
      The challenge is then extracting more data: whether it's a cartoon, a 3D rendered face, a sculpture an actual face.

      --
      im in ur .sig, writin ur memes.
    4. Re:Face detector by SquadBoy · · Score: 1

      That would pretty much be my point. The people who are claiming they can scan and then match faces to particular people. How the hell do we know that they can even pick out a face. Or in your point a real face.

      --

      Cypherpunks: Civil Liberty Through Complex Mathematics. Those who live by the sword die by the arrow.
  10. Interesting by Doesn't_Comment_Code · · Score: 5, Interesting

    ...while falsely identifying four objects as faces.

    That's interesting. The AI is sufficient to identify most faces. And it sees a few faces where none exists - not unlike people. Little kids point out when their bananas, carrots and peas line up just right to make a face. If they see it, why shouldn't a computer? What about the moon? Would this software see a face there? A man maybe?

    There is no point to this, just interesting thoughts that struck me while reading.

    --

    Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    1. Re: Interesting by Black+Parrot · · Score: 0, Troll


      "There is no point to this" ==> "(Score:5, Insightful)"

      Newbies should take note of this excellent karma-harvesting strategy.

      --
      Sheesh, evil *and* a jerk. -- Jade
    2. Re:Interesting by Anonymous Coward · · Score: 0

      Your last sentence should not contain the word "interesting", but rather "juvenile", "idiotic", "sophomoric", or "devoid of any real philosophical interest".

    3. Re:Interesting by Dr.+Evil · · Score: 1

      "Normal" infants recognize faces very quickly. It seems to be "instinct".

      I think this is a little fuzzier than most people make it out to be, and really odd if done in 2d still shots.

      If you show a computer a photo of some guy sitting down at a desk and ask "is this a face?", what you're really asking is... "is this a two-dimensional still projection of a three-dimensional space which contains a partial image of a face which is probably of an object which appears to be that of a human?"

      I think projection and interpretation of space is important, the photo on the guy's desk of his kids out at the beach would have to be discarded as an alternate scale where a giant replica office has been wrapped around an ocean setting, or both are valid and the office is suspended over a beach (load up the algorithms to identify this as an office, a beach, a window and then figure out if this is a resonable place to put a window in a building), the Dilbert strips would have to be discarded as "probably out of scale, probably not intended to resemble a human"

      I think you're absolutely right, spotting faces in shapes of cars, shapes of houses, cartoon strips, etc should not be completely discarded as "those aren't faces, the AI is wrong!", but instead discarded as "those are probably not intended to resemble humans"

      ...provided you're not taking the necessary shortcuts to get around limits of computational power.

    4. Re:Interesting by iabervon · · Score: 1

      For that matter, probably most people identify non-faces that look like faces by context. If you can tell there's no body attached, it's probably not a face. In order to figure that out, however, you need to understand things in front of other things, how much space a body takes up, how it bends, etc.

    5. Re:Interesting by Anonymous Coward · · Score: 0

      Your last sentence should not contain the word "interesting", but rather "juvenile", "idiotic", "sophomoric", or "devoid of any real philosophical interest".

      Hey, if you're not going to contribute positively, fuck off.

  11. Google Cache by Anonymous Coward · · Score: 1, Informative
  12. Gender Recognition by Black+Rabbit · · Score: 2, Interesting

    Can it differentiate between a male face and that of a female? Besides the obvious facial hair thing, what makes the two different anyway?

    1. Re:Gender Recognition by Anonymous Coward · · Score: 0

      There must be something, though, because some trannies without facial plastic surgery still look like guys.

    2. Re:Gender Recognition by acb · · Score: 1

      I suspect that the shape of the jaw and cheekbones would be the obvious thing. There are probably sets of measurements which can be plugged into a formula to give a maleness/femaleness index.

    3. Re:Gender Recognition by nule.org · · Score: 1

      The differences between an 'average' male face and an 'average' female face are less than the possible differences between two female faces or two male faces. There are lots of hints that a person takes into account when trying to determine what sex a person is, and very often it's things like fashion hints, posture and speech that we rely on. I'd be interested to see if a program based purely on static image recognition could do a very good job of guessing. Particularly if some of the typical hints were ruled out - hair length, facial hair, jewelry and makeup. The secondary sex characteristics that are built into the face that are considered more masculine are a heavier jaw-line, more prominent cheek bones and eyebrow ridges. But as I said before there's more variation in those features among women then there would be among an average woman and an average man. Interesting question you bring up.

    4. Re:Gender Recognition by Anonymous Coward · · Score: 0

      Case in point: Michael Jackson

    5. Re:Gender Recognition by UTPinky · · Score: 1

      Look at a picture of Michael Jackson. Can a human being even make this distinction?

      --
      I'm only paranoid because everyone is against me...
    6. Re:Gender Recognition by Jugalator · · Score: 1

      Male / Female Facial Differences

      While we are at it...

      Male / Female Skeletal Differences

      Obviously, the pelvis has a lot of differences. :-)

      --
      Beware: In C++, your friends can see your privates!
    7. Re:Gender Recognition by Jugalator · · Score: 1

      Btw, you gotta love their face examples and how they show DiCaprio is an example of a female face. :-)

      --
      Beware: In C++, your friends can see your privates!
    8. Re:Gender Recognition by Unregistered · · Score: 1

      female faces have tits underneath. Duh.

    9. Re:Gender Recognition by Anonymous Coward · · Score: 0

      Go ask your neighborhood transsexual. The male/female face differences can be a matter of life or death to an MtF or FtM.

      Brow bossing (men: distinct bony ridge above eyesockets, women: smooth).

      Eyebrow location and shape.

      Chin.

      Jaw shape.

      Hairline. Even before male pattern baldness sets in.

      Nose profile.

      Fat distribution.

      Those are the major ones. And yes, I'm a transsexual, so I'm constantly aware of this. Different ethnicities can be more 'male' or 'female' on average. And you get people with fairly androgynous faces who don't quite read as either at first glance. (I am lucky in that I started out androgynous!)

      There are other male/female cues, of course. Hand shape, trachea, and of course boobs. But really, your gender judgement is made almost entirely above the neck. Especially if you include the voice in that - I know an MtF who is damn good-looking, and passes as a woman easily, until she starts TALKING.

      -some genderqueer freak

    10. Re:Gender Recognition by Anonymous Coward · · Score: 0

      I am transsexual.

      I have seen bio-females with smaller breasts than I have, and bio-males with much bigger.

      So your observation is nothing more than sexist bullshit!

    11. Re:Gender Recognition by Anonymous Coward · · Score: 0

      the obvious: adam's apple

    12. Re:Gender Recognition by Anonymous Coward · · Score: 0

      Not all males have a prominent Adam's Apple.

    13. Re:Gender Recognition by Anonymous Coward · · Score: 0

      His face has been so fucked up by all his plastic surgery that he might as well be wearing a latex mask. Hardly a typical face of either gender.

  13. Give Me a Break by Anonymous Coward · · Score: 2, Insightful

    I can't tell people apart sometimes. I have a VERY hard time remembering someone's name, especially when they're Person #12 I Met On The Tour Of The Office out of The 37 People I Met On My First Day.

    But am I afflicted with "Face Blindness"? NO! I have a shitty memory for faces, and that's it! I don't have some made-up malady that can be cured with thousands of dollars of useless medication advertised on TV!

    The technology sounds cool. The culture of euphemism in the US just pisses me off, that's all.

    1. Re:Give Me a Break by Anonymous Coward · · Score: 1, Insightful

      This technology was not made to help individuals; it's a "security" product. It's for picking terrorists and other criminals out of a crowd. At least that's how it's going to be marketed to the government by the developers. And maybe if it worked as advertised, it wouldn't piss so many people off, but it just makes too many mistakes. Of course, this is supposed to be the new, improved version *rolls eyes*.

    2. Re:Give Me a Break by FictionPimp · · Score: 1, Troll

      I have the exact same problem, but its called "I dont give a shitidius"

      Basically, I dont give a shit about 99% of people, so I only remember 1% of the people i meet.

    3. Re:Give Me a Break by Anonymous Coward · · Score: 0

      Just remember that 50% of the people you know are below average.

    4. Re:Give Me a Break by egomaniac · · Score: 4, Insightful

      I don't have full-blown face blindness, but I'm not terribly far away.

      If two people look vaguely similar, I have tremendous difficulty telling them apart. I don't mean "they could be twins" similar either -- even "they could be second cousins" can seriously throw me off. I can meet a person, spend twenty minutes talking to that person, walk away and come back, and be completely unable to pick that person out of a small crowd. Depending upon how "average" the person's face was, I might not even experience any sense of familiarity at all when looking at them.

      You learn other tricks for recognizing people after a while. I usually note what people are wearing, so that if I run into them again the same day I have a good shot at recognizing them. Hair color and style help a lot, and tend to remain constant for a substantial time. I'm also very good at identifying voices, so I often wait to hear a person speak before I feel confident that I have correctly identified them. I also rely on my wife to help me remember people a lot -- fortunately she's healthy in this regard, and very understanding of the difficulty I have.

      To give you an idea of how bad things are, a long time ago I was away on business for two months. My girlfriend (now my wife) and I had been together for two years at that point, but I hadn't seen her during those two months.

      She had changed her hairstyle while I was away, so when I got off the plane I didn't recognize her. I noticed this girl smiling at me, and I thought she looked sort of like my girlfriend, but it wasn't until I was within five feet of her that I was sure it was her.

      Can you imagine dating a girl for two years, and then having trouble recognizing her after a mere two-month absence? And I don't have face blindness. I just have moderate difficulty identifying people, compared to the full-blown disorder.

      So, yes, call it a "culture of euphamism" all you like, but I certainly believe that this is a real disability that affects real people.

      --
      ZFS: because love is never having to say fsck
    5. Re:Give Me a Break by Anonymous Coward · · Score: 0

      Bah, i'm not trolling, i'm serious!

    6. Re:Give Me a Break by micromoog · · Score: 1
      but it wasn't until I was within five feet of her that I was sure it was her.

      I assume you've ruled out nearsightedness?

    7. Re:Give Me a Break by egomaniac · · Score: 1

      I assume you've ruled out nearsightedness?

      I am very nearsighted, actually, but I wear glasses. My vision is corrected to 20/20.

      --
      ZFS: because love is never having to say fsck
  14. Conspiracy time by Leffe · · Score: 1

    # Example: http://us.news2.yimg.com/us.yimg.com/p/nm/20010706 /mdf21464.jpg
    # Supported image formats: jpeg, gif, tiff
    # Please do not submit any images that are larger than 2.5MB
    # Please do not submit any images that have pornographic content

    Why would a machine care about that... oh, wait, maybe there's a human doing it all!!!

    1. Re:Conspiracy time by mobby_6kl · · Score: 1

      Anybody tried submitting the goatse pic? :D

    2. Re:Conspiracy time by Slick_Snake · · Score: 1

      There may be workers that filter through the images to help with the training of the system. After all how is it going to learn unless there is a mechanism that detects right and wrong responces.

    3. Re:Conspiracy time by JUSTONEMORELATTE · · Score: 1
      There may be workers that filter through the images to help with the training of the system. After all how is it going to learn unless there is a mechanism that detects right and wrong responces.
      Reasonable at this stage of the project. But why not just spend an afternoon with Google's Image Search until you get a good query string? Presto! You've got a corpus of images that will keep your human verifier busy for the rest of the year.

      --
    4. Re:Conspiracy time by emtechs · · Score: 1
      More likely they are concerned about having a huge store of porn in their database.
      Never know who's rules you are breaking these days.

  15. NASA needs this by IDigUNIX · · Score: 1

    ...they can have it analyze all of the Mars orbital images, and then choose landing spots for future probes that are optimally distant from any "facial anomolies". That way they can avoid having to investigate them, and keep the tin-foil-hat crowd stocked with fresh conspiracy fodder.

  16. Quick, someone send... by Caeda · · Score: 0, Funny

    Someone send two photo's of George Bush. One of his face, one of his ass. Its time to prove even our most sophisticated computer can't tell the difference. ;)

    --
    ~~ Please keep your arms, legs, and outright stupidity inside the ride at all times. Thank You ~~
    1. Re:Quick, someone send... by carvalhao · · Score: 1

      Hey, that's not a fair trial... the larger holes in each picture expell the same kind of substance! Way to hard even for a human!

    2. Re:Quick, someone send... by pubjames · · Score: 2, Insightful

      Actually, that raises an important point. This system can identify a face, but does it identify only human faces? After all, chimpansees have many of the same facial characteristics as some humans.

    3. Re:Quick, someone send... by Caeda · · Score: 1

      Your right. I don't think I could subject even a computer to that kind of torture. Might get stuck in a infinite loop. Perhaps it'd just explode... then again, it might just take days to determine both pictures seem identical :)

      --
      ~~ Please keep your arms, legs, and outright stupidity inside the ride at all times. Thank You ~~
    4. Re:Quick, someone send... by LiquidCoooled · · Score: 1

      somebody sent a load of pics regarding bush and the chimp!

      here

      Surprisingly, it did quite well!

      --
      liqbase :: faster than paper
    5. Re:Quick, someone send... by MattAustin · · Score: 1

      I don't think it can identify non-human faces. If you look at the results from May 26, 2004 (Submission #23) you'll see that someone uploaded an image from some sci-fi show. The 2 humans are identified, but the other guy is not.

    6. Re:Quick, someone send... by basic70 · · Score: 1

      Please, let me moderate moderations!

      Score Insightful? For that link? Did you check the pictures?

    7. Re:Quick, someone send... by MattAustin · · Score: 1
      Someone already tried that Bush or chimp image. The algorithm actually did rather well. It identified all of the faces of Bush and even a couple of the chimp faces.

      The interesting thing about the chimp faces was that the area it identified was usually a smaller region of the chimp's face that it just mistook for a face.

      -Matt

    8. Re:Quick, someone send... by Anonymous Coward · · Score: 0

      I'm assuming that result will be posted sometime after May 26th 2004?

    9. Re:Quick, someone send... by MattAustin · · Score: 1
      Oops, my mistake. May 6th, I meant. Here's the link to the scene from Farscape .

      -Matt

  17. Re:I don't have two eyes you insensitive clod by NanoGator · · Score: 2, Funny

    "Or a mouth, or a nose, or two ears. I'm the man without a face. "

    So your only distinguishing feature is a large crack?

    --
    "Derp de derp."
  18. Red eye by Todd+Fisher · · Score: 0, Offtopic

    Source Kodak:

    The "red-eye" phenomenon has been familiar to photographers since the introduction of synchronized flash picture-taking. It's caused by the reflection of light off the blood vessels of the retina of the subject's eyes. It occurs most often when the flash is located close to the picture-taking lens of the camera.

    The red-eye effect tends to be more evident when the subject is young and has blue or gray eyes which reflect more light than darker eyes. Also, children have larger pupils and less pigmentation than adults' eyes and they transmit more light back to the camera lens.

    Red-eye wasn't so common with older, bulkier cameras that had separate flash units. These units were attached to a handle or flash bracket several inches away from the lens, or the photographer could detach the flash and hold it away from the camera. With today's popular small cameras with built-in flash, the flash is closer to the lens. However, this doesn't mean the red-eye is unavoidable.

    The following techniques can help reduce red-eye:

    Increase the level of light in the room by turning on all the room lights. The added light will cause the subject's pupils to contract, reducing the reflective surface that causes red reflections.

    - Have the subject look at a bright light (for example, a room lamp or a ceiling light) just before you take the flash picture. The bright light will reduce the size of the subject's pupils

    - Have your subject look slightly away from the camera lens rather than directly toward it. This will help reduce or eliminate the reflections that cause red-eye.

    - If your camera has a detachable flash, move the flash away from the camera lens. You can attach the flash to the camera with a flash cord and handhold it or clamp it to a nearby object.

    --


    --I'm not talking about dance lessons. I'm talking about putting a brick through the other guy's windshield.-
  19. another study says.. by KReilly · · Score: 4, Funny

    face-blindness has a direct correlation to breast size...

    1. Re:another study says.. by igny · · Score: 1

      Does it have any connection to blondes?

      --
      In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
  20. Yeah, cool tech, but how will it be used? by Kegster · · Score: 1

    As a universal panacea to protect all right-thinking people from terrorists, anti-capitalists and whatever else the state and the corps that support it decide is unacceptable behaviour for consumers?

    David Blunkett(UK Home Secretary) will love it, despite the fact that it will be pretty much useless for preventing anything, unless you have a universal database of the facial characteristics of everyone living in or entering the country (through illegal as well as legal channels).

  21. MED Award by sssmashy · · Score: 5, Funny

    Though face detection is easy for most people, some suffer a perplexing disorder called face blindness, or prosopagnosia, which is an inability to discern the differences between faces...One such sufferer, who is part of a research study by Behrmann, can't recognize his own children when he picks them up from day care. He relies on the day care workers to bring his children to him; failing that, he carries a "cheat sheet" of photographs that can help him make out who's who.

    We just found a candidate for the Most Embarassing Dad of the Year Award!

    Dad: Hi, I'm here to pick up my son, Billy.

    Day Care Worker: Sure, which one is he?

    Dad: Uhhh... (pulls out photograph) I think he's the one one the left... no wait, in the middle. I'm not really sure.

    Day Care Worker: Uh, OK, sir, whatever you say. Let me just leave the room and, uh, get Billy. (leaves room, dials 9-1-1. A few minutes pass)

    Kid at Daycare: Hey Billy! The cops are arresting your dad again!

    1. Re:MED Award by physick · · Score: 4, Interesting

      Almost all human faces have common characteristics, such as two eyes and one mouth.

      The "almost" reminds me of the joke: "Do you know that you have more than the average number of legs?

      Some people have lost one or both legs, but no one has three or more. So the average number of legs is slightly less than two."

    2. Re:MED Award by Anonymous Coward · · Score: 0
      What tripped me up was the next sentence: Almost all human faces have common characteristics, such as two eyes and one mouth. Still, some people, affected by face blindness, cannot recognize one face from another one.

      Still? That implies difficulty in spite of something that makes it easier. I'd think it'd be easier to tell people apart if everyone had a different number of eyes and mouths.

    3. Re:MED Award by greylouser · · Score: 1
      \acknowledges humor

      \offtopic = 1

      \nitpick = 1

      To be fair, you should first specify that, as an average, you're referring to the mean number of legs rather than the mode.

      \nitpick = 0

      I know. I'm seeking help.

  22. A Good Tool for WOT by USAPatriot · · Score: 2, Interesting
    I'm surprised this technology isn't in more widespread use today. Casinos are known to have implemented face-recognition technology to recognize cheater and card counters and bar them.

    If something like this were installed in airports, bus terminals, landmarks, and other public places, we could have a very effective way of stopping potential suicide bombers, terrorists or other evildoers in their tracks. What if the video camera that captured Mohammed Atta had been linked to face-recognition software that had his picture in a database? 9/11 could have been stopped right there.

    I would rather put my trust in this than some rent-a-cop making minimum wage to spot suspicious people. It's been reported how unreliable eyewitness testimony is, this technology would make a much better crime fighting tool.

    --

    Slashdot Moderation: From positive to terrible in 2 "insightful" posts.

    1. Re:A Good Tool for WOT by xystren · · Score: 2, Interesting
      If something like this were installed in airports, bus terminals, landmarks, and other public places, we could have a very effective way of stopping potential suicide bombers, terrorists or other evildoers in their tracks.

      But at the bolded places now, they are already doing this (perhaps with out the face recognition). Ever tried to go to the Sears or the CN Tower or even a ride at an amusement park where they have there "let's take a picture of you, so we can try and sell it to you" after your tour/trip/ride???

      Next question, have you ever tried to *NOT* have it taken? I pulled out a pic of myself and g/f at the Sears Tower and told them, 'No, I'm not interested in one, I already have one, just let me through..', and it was "Uhh, sir, it's policy, we have to photo everyone going through."

      Hmmmm, I wonder why? If it's just a marketing/$$$ game, then why would the waste the resources to print out my photos when I've stated I won't be buying another one (and even produced a past one to support that argument).

      Perhaps I'm a bit paranoid, I would suspect that there is *more* than just a "let's try and sell a tourist a picture" type thing. --- We can't go on like this, with suspicious minds...
  23. Conspiracy? by baudilus · · Score: 4, Funny
    ...you can submit an image for analysis and receive the results by e-mail.

    So now they can link my face to my e-mail address? No thanks.
  24. well.. by icekillis · · Score: 0

    It fails in the most important thing: IT doesnt' tell you if the subject is at least 18 years old!

  25. It'll break! by Doesn't_Comment_Code · · Score: 3, Funny


    I bet it breaks if you pass it a test image of Michael Jackson.

    --

    Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    1. Re:It'll break! by Anonymous Coward · · Score: 0
      I bet it breaks if you pass it a test image of Michael Jackson.

      That's not a fair test. Jackson no longer has a face.
    2. Re:It'll break! by Unregistered · · Score: 1

      Yea, it only finds the 8 year old next to him

  26. I used it in a project two years ago by Anonymous Coward · · Score: 1, Interesting

    I used this library for a multimedia installation two years ago. But the source was not available... I still managed to integrate it in my project, and I had a lot of fun. I should build a web site to document this project.

  27. Re: Okay, call me crazy by jwcorder · · Score: 1

    "Still, some people, affected by face blindness, cannot recognize one face from another one."

    Ok call me crazy, but what the hell is face blindness. Unless you are talking about someone is truly blind, I cannot imagine an actual physical defect where people cannot see the difference between let's say your mommie's face and the face of Bill Gates.

    I say instead of creating wonderful pieces of technology, just get the damn kid a pair of glasses or maybe some Prozac for his mental illness.

    --
    http://jayceecorder.blogspot.com
  28. It's Tinfoil Time! by WormholeFiend · · Score: 0

    "The Face Detector Demo is available online and you can submit an image for analysis and receive the results by e-mail."

    1. Collect face database
    2. Collect "live" email database
    3. Sell databases (no "?s" about it)
    4. Profit!

    1. Re:It's Tinfoil Time! by Anonymous Coward · · Score: 0

      I'm not sure it qualifies as a tin foil type plan unless one of the steps is unexplainable, hence the ???? convention before the 'profit' step. What you have seems like an entirely workable and deplorable business strategy. Coming soon to a mall near you...

    2. Re:It's Tinfoil Time! by WormholeFiend · · Score: 1

      what if the ??? are put as

      5. ???

  29. Only to recognize where faces are...but good by icekillis · · Score: 5, Interesting

    The algorithm only recognizes where the face is. It does NOT recognize the face to match it with another picture.
    The algorithm is almost scary, watch this sample
    http://vasc.ri.cmu.edu/demos/faceindex/05062004/us ers/2236.html

    The problem is that even if you can recognize where this 80 pixel face is, it will be very hard to match it up against features of known people. Several [automated] face recognition systems implemented in Florida failed. In more than 3? months they failed to identify a single known offender.

    1. Re:Only to recognize where faces are...but good by MellowTigger · · Score: 1

      I'm glad that somebody else identified this distinction. Persons with prosopagnosia ("face blindness") are unable to identify a particular face. They are perfectly capable of identifying where a face is; they just don't know whose name that face belongs to.

  30. Corny Joke Alert by juggaleaux · · Score: 0, Offtopic

    Face it, this is great news.

    1. Re:Corny Joke Alert by Anonymous Coward · · Score: 0

      Woah, I was blind sided by that one.

      Perhaps we should have a face off?

      Ugh, I'm not good with face puns.

  31. I wonder if... by Anonymous Coward · · Score: 0

    it would detect famous slashdot trolls as faces, i.e. the infamous goatse.cx

  32. Maybe it's monday, but by Prince+Vegeta+SSJ4 · · Score: 2
    I can't quite figure out this statement:

    By analyzing only 768 pixels, the system can detect 93 percent of the faces in a set of images while falsely identifying four objects as faces.

    Is it - four objects in a specific set of images, or four specific objects.

    for example

    • all 'insert object here' objects are mistaken for faces
    • or is it only four random objects in this specific set
    1. Re:Maybe it's monday, but by Anonymous Coward · · Score: 1, Informative

      It's the latter. Except it's probably not really "random" - there are, of course, reasons that those 4 gave a false +. It's just that we don't know why (unless you consider ``the algorithm is not sufficiently advanced'' to be an explanation, which I don't).
      On another note, it is a very stupid writeup which gives a ratio for correctness and an absolute value for the number of type I error, without even giving the size of the dataset. For shame...

    2. Re:Maybe it's monday, but by Frett2 · · Score: 2, Informative

      The dataset they are probably using is the CMU+MIT hard face detection dataset available here (link to 30 meg tar file at bottom of page): http://vasc.ri.cmu.edu/idb/html/face/frontal_image s/ This dataset is made of images that contain static, contain faces at unique angles, contain complex scenery which would mess up the way most face detection algorithms work, and more "hard" cases. It's the dataset that all face detection programs are tested against. If I remember correctly most other face detection programs only achieve approixmatly 80-85% detection rate when they only get 4 false positives in the results, so this algorithm is definatly an improvement.

  33. Does it ever get... by Anonymous Coward · · Score: 0

    ...the beer goggles?

  34. Re: Okay, call me crazy by nathan+s · · Score: 2, Interesting

    I don't believe I have actual face blindness, but I definitely suffer from face-recognition problems. It has been one of the defining factors of my personality, to the point that most of my friends know not to be offended if I do not recognize them until they actually speak to me. Interestingly enough, my face recognition is quite good within small time periods (1 or 2 days) but degrades quite quickly; if I haven't seen you in a week or two, then no matter how long I've known you, odds are that I won't recognize you if we meet on the street or in a small crowd of people.

    This also goes along with an inability to remember faces; I can't remember at this moment what my girlfriend's face looks like, even tho I saw her a few hours ago. Hell, I can't even remember my family's faces, and I spent years around them.

    Just amusing thoughts from experience..

  35. Identical Twins and Faces by Dareth · · Score: 1

    I have a pair of sisters who are identical twins.
    How well can the system differentiate between them?

    It might be easier as the parent referred.. think they are opposite end identical... one got a face like a but and the other got a but like a face!

    --

    I only look human.
    My mother is a halfling and my dad is an ogre, so that makes me an Ogreling
  36. jesus in clouds by morcheeba · · Score: 1

    So, if this picture is detected as a face, does that count as correct, or not? I submitted it, so we'll see what they get.

  37. Previous submissions by TummyX · · Score: 1

    I see they have a list of previous submissions. I wonder how long it'll be before someone submits goatse man.

    1. Re:Previous submissions by tehcyder · · Score: 1

      oops ;-)

      --
      To have a right to do a thing is not at all the same as to be right in doing it
    2. Re:Previous submissions by Tablizer · · Score: 2, Funny

      The worse insult in the world would be this machine matching your face up with a goatse image.

  38. Missing Option on form by sboyko · · Score: 1

    Email address: bob@yahoo.com
    URL: www.geocities.com/bob/me.jpg
    Are you a criminal? _____

    --
    SCO, Microsoft, P2P, what's your hot button?
  39. That'd be a trip... by Greyfox · · Score: 4, Funny

    Walk up to someone you never met before, greet him by name, and ask him about all the various details of his life are going that you shouldn't know because you never met him before. It'd be a good way to freak people out, especially of your borg implants aren't particular noticable...

    --

    I'm trying to teach myself to set people on fire with my mind... Is it hot in here?

    1. Re:That'd be a trip... by ahem · · Score: 1

      This recalls the scene from Snow Crash where Hiro has gone gargoyle and approaches the motorcycle dealership's salesman with just just a line of patter. Loved it.

      --
      Not A Sig
    2. Re:That'd be a trip... by oaklid · · Score: 1

      No implants required. Simply read the weblogs of strangers in your city and keep an eye peeled. (Recognising them is easier if they post photos.)

      It happened to me. http://www.mediatinker.com/blog/archives/008063.ht ml

    3. Re:That'd be a trip... by subtropolis · · Score: 1

      In many places, it's distinctly *not* a good idea to have the locals believe you're a 'spy'. A knee-capping (if you're lucky) will really fuck up your week.

      --
      "Our interests are to see if we can't scale it up to something more exciting," he said.
  40. Re: Okay, call me crazy by Anonymous Coward · · Score: 1, Informative

    You're crazy.

    Now that we have that out of the way, "face blindness" is a mental condition where the part of the brain that normally remembers what a face looks like does not work properly (somewhat like a learning disability). It has nothing to do with the person's sight, but with their memory.

    It's rare, and quite odd, but it's recognized as an actual disorder.

  41. Re: Okay, call me crazy by jwcorder · · Score: 2, Informative

    I googled the topic and found some really amazing information. This site covers this topic greatly. I was amazed at this illness. I am no were near a great person at remembering faces but I think it's amazing that if I walked up to you and said hello, then walked away for 10 mins and came back you wouldn't recognize me. It's like Finding Nemo all over again. :)

    --
    http://jayceecorder.blogspot.com
  42. Re: Okay, call me crazy by richg74 · · Score: 4, Interesting
    There is an actual mental handicap, called prosopagnosia, which is marked by the inability of the person to identify individual faces. That is, the person can recognize that what (s)he is looking at is a face, but not whose face it is.

    Steven Pinker talks about this in his book, How the Mind Works (New York: W.W. Norton & Co., 1997). He writes:

    Many psychologists believe that face recognition is special. In a social species like ours, faces are so important that natural selection gave us a processor that registers the kinds of geometric contours and ratios needed to tell them apart. Babies lock onto facelike patterns, but not onto other complex and symtrical arrangements. when they are only thirty minutes old.
    He also discusses a patient, LH, who was unable to recognize faces following a severe head injury, although he was in other ways entirely normal.

    It's important to note that this is a different question than the one the software addresses: it tries to distinguish which images are faces and which are not, not whose faces they are.

  43. another similar project by arcanumas · · Score: 1

    Not the first attempt at the problem:
    http://franck.fleurey.free.fr/FaceDetection/index. htm

    --
    Slashdot Sig. version 0.1alpha. Use at your own risk.
  44. Gender and race show the limits of this (and us) by ianscot · · Score: 4, Insightful
    The subtle stuff that tells you which faces are male and female doesn't have an objective truth behind it. There is no concrete set of criteria you can rely on in making those judgments, it's just a set of rough truths you work with to get by. You can't always make a good guess, there are borderline cases: to use your own intended-to-be-obvious example, some women have lots more facial hair than others.

    There's no way for this process to reliably determine something like race, either -- not that doing so is that desirable anyway. The characteristics that make up "an African American guy" are just not nearly as concrete as we think they are from day to day. I have a neighbor who thinks all the Somali people in my area are "Arabs." Her category is a little too broad. It seems to me like she's forcing certain expressions onto their faces, too, as part of her image of what "Arabs" are like.

    People's minds love to categorize. Sometimes, a lot of the time, we force information into categories it doesn't quite fit. (Refer to: State Department intelligence from Iraqi exile organizations.) Even when the information is essentially noise, we try to sort it and sift it. As a result we persist in holding weird ideas: astrology, because the paper tells us something vague and we run the events we see past that filter.

    We should expect our tools to share some of those biases and blind spots. As much as we might try to address that, we have the blind spots ourselves, so it's hard to know how to counter the problem.

    --
    "Fundamentalism" isn't about divine morality. It's about human authority.
  45. "Face Blindness" by raygundan · · Score: 4, Informative

    While I agree with you about the general "culture of euphemism," as you put it-- I don't think this is one of those times. Face Blindness is not referring to people like you and me who are just lousy at remembering who we met, but rather people with profound neurological disorders who *literally* cannot tell a face from something vaguely facelike, like a vase or a particular arrangement of shadows. This goes far beyond not remembering the guy you met at a convention a year ago-- but rather not even being able to tell the difference between his face and the PDA he was holding.

    For a quick read on it, check out The Man Who Mistook His Wife for a Hat. The things that happen to the poor people in this book as a result of disease, physical damage to the brain, or conditions they were born with are bizarre but definitely interesting.

    1. Re:"Face Blindness" by AndroidCat · · Score: 1

      I thought of that too--then I checked to see if anyone else had mentioned it. Rather waste a search, here's another good link.

      --
      One line blog. I hear that they're called Twitters now.
    2. Re:"Face Blindness" by raygundan · · Score: 1

      Funny coincidence with your .sig-- I read the book I mentioned as part of a science fiction literature class at Purdue years ago, as preparation for discussion of Phillip K. Dick's "A Scanner Darkly."

    3. Re:"Face Blindness" by AndroidCat · · Score: 1

      But will we recognize Phillip K. Dick's work in the coming movie? :)

      --
      One line blog. I hear that they're called Twitters now.
    4. Re:"Face Blindness" by raygundan · · Score: 1

      That's about as likely as me sprouting wings and flying off to the moon for a nice afternoon of cheese-tasting.

      Of course, Lord of the Rings turned out okay, and boy did I enjoy the moon-cheese that day. So it has now been conclusively proven that movies based on books don't *have* to ruin the story, just that it's overwhelmingly likely.

  46. freak fest by Anonymous Coward · · Score: 0

    It sure as heckfire will be fun to watch all the slashdotters pics tomorrow.

  47. They're not talking about you by MacGabhain · · Score: 1
    Face Blindness is an actual condition, not just a shitty memory for names. Recognizing other people's faces is a big, very early step in the evolution of humans as social creatures, and it's an ability that's severely stunted in some people's brains.

    An example of the difference: I teach at a community college. I have a poor memory for names. By the end of the term I have trouble remembering more than a couple dozen of my 120 or so students' names. Someone with face blindness wouldn't be able to recognize which ones were in his classes when walking around campus.

  48. An online-DEMO of some NEWER stuff by ControlFreal · · Score: 4, Interesting

    Seriously, that's old. I'm a computer vision Ph.D. student, and there now are much faster methods. I'll just refer to my old comment.

    A demo can be found here. You can contact me for more details...

    Current really fast methods use cascades of very simple classifier that are very weak themselves, but very strong when combined. The work of Viola & Jones is what most of the stuff is centered around nowadays.

    Do your own here:

    http://argus.cs.unimaas.nl/fddemo

    --
    Support a Europe-related section on Slashdot!
    1. Re:An online-DEMO of some NEWER stuff by jumpfroggy · · Score: 1
      For other people interested in playing around with Face detection, you can take a look at OpenCV ( http://www.intel.com/research/mrl/research/opencv/ ). It's a general purpose image processing library, but it also has some classifiers. I'm using it right now with a webcam to detect faces for some research I'm working on, using the HAAR classifiers. Not too robust, but interesting. It detects my eyes using small scaled subsections, then tracks them using object tracking methods.

      It's useful for a lot of things. Too quickly we think that it not useful if it can't recognize individual faces. But I'm using it for research, hopefully ending up in a kiosk. If a kiosk knows if there's a person (doesn't matter who, specifically) in front of it, it can respond more intelligently. There are lots of applications for "there's a face here", and not just "john's face is here".

    2. Re:An online-DEMO of some NEWER stuff by Nova77 · · Score: 1

      Yeah. The Viola & Jone's paper is the best. I am working on a real time detector right now for an artistic festival. The core is already programmed and it can classify very well and very fast. The hardest work was to build a database! :P
      Expect a sourceforge code soon! :)

    3. Re:An online-DEMO of some NEWER stuff by ControlFreal · · Score: 1

      Good! It's good to have as much of this stuff as possible in Open-Source form. I think it should be possible to make the Viola-Jones framework much faster by using a number of tricks:

      • First, don't program this stuff in C++ with all kinds of function calls: do it inline, and in C.
      • Use SSE(2) instructions to speed up wavelet calculation, and REUSE wavelets for adjacent windows. (SSE(2) could be used to calculate multiple wavelets in one go).
      • Use some kind of cache-optimization, like ATLAS does for LAPACK.

      I didn't implement any of this, as I'm working on context-based detection, which in my case is just a proof of concept. That means actual speed is not so much of a concern. However, I'd be very interested in seeing some code ;)

      --
      Support a Europe-related section on Slashdot!
    4. Re:An online-DEMO of some NEWER stuff by ControlFreal · · Score: 1

      Interesting! Do you have any detection-results on a dataset, and/or a ROC-curve (detection rate vs. false-positive rate)?

      Did you use the pretrained classifier as well? And what window-size do you use?

      --
      Support a Europe-related section on Slashdot!
    5. Re:An online-DEMO of some NEWER stuff by Nova77 · · Score: 1
      > First, don't program this stuff in C++ with all kinds of function calls: do it inline, and in C.

      I am programming in C++ with but with a lot of inline stuffs when using the wavelets features. I don't think that using C++ in this case will have a big impact.

      > Use SSE(2) instructions to speed up wavelet

      At the moment the implementation is plain, but this is surely a good idea. In fact I was even thinking of using GPU code for that.

      > REUSE wavelets for adjacent windows.

      I don't see the advantage in doing that. Isn't going to be the same value of the haar feature?

      > Use some kind of cache-optimization, like ATLAS does for LAPACK.

      That's also a good idea. Actually I was thinking to look at the code of Blitz++ for some suggestions (it uses a lot of templates tricks for cache-optimization).

      You will see the code in about a month, I guess. First we want to clean it up and have something working (there are already too many empty project in sourceforge! :) ).

  49. Re:jesus? by Anonymous Coward · · Score: 0

    i thought it looked like charlie manson;-)

  50. Re: Okay, call me crazy by acb · · Score: 4, Interesting

    The human brain is not a general-purpose computer. There are specialised modules, which evolved by natural selection, for various tasks which were evolutionarily adaptive in the ancestral environment. Which is to say, if being able to perform one type of mental processing quickly helped your hunter-gatherer ancestors survive, find fit mates and not get cheated or otherwise bested by competitors, it gradually evolved into an optimised piece of neural hardware, its template coded in their DNA. Being able to recognise faces quickly (and thus be able to match an image of someone to what you know about them and their reputation) was a major advantage in a highly social environment, and so evolved into a highly optimised module.

    Sometimes, through various disorders, these modules don't work properly. Which is why conditions such as autism (dysfunction of the relating-to-other-people module), schizophrenia (inability to distinguish between internal and external stimuli), face blindness (the face-recognition module). and so on, can exist. Sure, a face-blind person can step themselves through a face-recognition algorithm, but it's slow and laborious, and by the time you're done, that cute girl/guy you're wondering whether you recognised has moved on.

  51. Let me explain... by fingerfucker · · Score: 5, Informative

    I am no expert in this technology, but I am somewhat knowledgable about it, let me explain something.

    You won't understand how hard is it to actually pull off something like face recognition until you yourself actually sit down and try it, only to realize that the problem is much more complex to solve when it has to be so all-encompasing.

    The first step to face recognition is to recognize where the face is. The result of this process are quadrilaterals that carve out the face so that when you crop, you are left with exactly the face (frontal, or profile view or other).

    A common technique used to do that is to locate the eyes. Most faces (heck, even those with veils on them for relegious reasons!) will contain eyes. Then, when detecting where the face is, you are only left with not having covered people who are wearing sunglasses (which are much easier to detect).

    After you have located the eyes, you gauge by their proportions the approximate proportions of the face. Then, you apply an iterative technique (varies in principle, typically based on differential calculus combined with numerical methods of approximation) to locate the bounds of the face so you can eventually crop it to know WHERE THE FACE IS.

    "Obviously", the iterative technique has to be able to detect false positives via a threshold set that will rule out the non-face. However, once you have located the eyes with certain reliability, the overall chance that you have come across a face is pretty solid.

    The problem is complicated as it is already as you can see!!

    Only after FINDING the face, you can start MATCHING the face. At that point you are facing a number of problems that the imagination of most /.-ers can conceive of... Bierds, smiles, teeth-showing, frowns, skin tone changes and the most popular by all scientist: plastic surgery....

    A common approach to the actual face matching is a technique of the so-called eigenfaces, whereby you compute a "common" face of the pool and then you can navigate down the specialization of characteristics (e.g. bigger, bigger, bigger nostrils) as you drill down, narrowing down the pool of possible faces.

    There is nothing that takes away from how much state-of-the-art CMU's research is. It would be like saying "why is someone dealing with virtual memory management of an operating system if by now, we already have user applications for the OS". Do you see the flaw in such thinking?

    The science behind is a lot of mathematics, so dear parent, please don't be ignorant of this type of work just because you don't understand its complexities...

    1. Re:Let me explain... by Anonymous Coward · · Score: 0

      I couldn't have said it better.

    2. Re:Let me explain... by mike3411 · · Score: 2, Funny

      this is really a great response, especially from someone whose username is "fingerfucker" =)
      kudos!

      --
      Mod me down, and I will become more powerful than you can possibly imagine!
    3. Re:Let me explain... by AbstracTus · · Score: 1

      "...quadrilaterals that carve out the face so that when you crop, you are left with exactly the face.." Sounds like a nasty procedure

    4. Re:Let me explain... by subtropolis · · Score: 1

      I second mike3411's sentiment. It really looks like everyone hear thought this was about putting a name to a face, rather than just finding the fucking thing first.

      While we're on-topic: any idea of related softwares out there for playing with? I'm more interested in *shape* detecting, esp. that of the environment around the camera. I came across a site awhile back which discussed techniques for finding the vertices of a room, corridor, etc. and deducing the location based on stored maps. I wish i could find it now.

      --
      "Our interests are to see if we can't scale it up to something more exciting," he said.
  52. Re: Okay, call me crazy by Caeda · · Score: 3, Informative

    The difference, is something I've had to deal with at various jobs. Although I can after seeing a person for quite some time (several hours or meeting/talking with them 5-6 times for an hour or more) I can keep their face and name in memory, I cannot do this on a short term basis. Example. I've worked in a grocery store and a K-Mart. People will often stop you to ask where something is, and, if the item is out, you have to go check in the back for them. When people would ask me for something, and I'd have to go to the back, I'd have a problem finding them if they moved from the time I left to the time I came back. Half the time, I would be lucky enough that either the person didnt move, was waiting for me and spoke up, or was wearing some ungodly horrible color of clothing that I could instantly identify and find. The other half of the time, the person was wearing something dull and wandered off a little way and I would have no chance to recognize them unless they came back to me. Really, its a pain in the ass to be that way, as your always wondering if your going to find the right person when you get back from something. But its nothing you cant live with... I'd imagine it'd be horrible for it to get worse and not recognize family...

    --
    ~~ Please keep your arms, legs, and outright stupidity inside the ride at all times. Thank You ~~
  53. File this under 'why is this news?' by dot+niet · · Score: 0

    http://www.identix.com http://www.viisage.com http://been.done.before

  54. Ugh broken logic by Elwood+P+Dowd · · Score: 3, Insightful

    Still, some people, affected by face blindness, cannot recognize one face from another one. So it's understandable that face recognition is a major challenge for computer vision systems.

    Face blindness just shows us that the specialized hardware we have for face recognition is so incredibly accurate that we rely on it completely and have no alternate methods of face recognition. When it's broken, other parts of our brain don't step in either because a) it's a hard task or b) they just don't have access to the relevant visual information. Face recognition could be totally simple, if this were our only measure.

    --

    There are no trails. There are no trees out here.
    1. Re:Ugh broken logic by ZorbaTHut · · Score: 1

      Still broken logic . . . as far as we know, humans have half a dozen methods for recognizing faces, all of which work flawlessly - and when someone has "face blindness", the problem isn't in the recognizing, it's in the chunk of the brain that processes and deals with that information.

      Of course we don't really have any way of knowing. :)

      --
      Breaking Into the Industry - A development log about starting a game studio.
    2. Re:Ugh broken logic by Elwood+P+Dowd · · Score: 1

      We kindof do. Have a way of knowing that particular thing. The part of the brain that deals with recognizing faces in the visual field is pretty well identified. Face blindness happens when that part of your brain is damaged.

      If there's damage before or after that spot in the path between light coming in your eyes and your mouth saying "I see a face" then other functions will also be impaired. The fact that we can't recover from that particular kind of damage is notable, too. We can't use our general-purpose object identification hardware to identify faces. That could say all kinds of things. I don't know about face-blindness and it's characteristics to tell you exactly what it says, but it certainly doesn't prove that face recognition is hard.

      --

      There are no trails. There are no trees out here.
  55. open source/academic projects? by d4rkmoon · · Score: 2, Interesting

    I wonder if there are open-source/academic projects that are in relation to this. I've been testing GSPY and some other security camera software as of late. If you could do testing on this type of software and coincide with facial recog, there could be a lot of useful things that have nothin to do with homeland security and the like. (such as having a computer system activate off standby from facial recognition at a certain point). So I like science fiction... don't we all.

    --
    -- Friends don't let friends buy Nokia.
    1. Re:open source/academic projects? by arcmay · · Score: 2, Informative
      Most of this research falls into two categories: Government-funded work at universities, and private research by companies looking to sell a commercial product. While it is near impossible to get developers on commercial systems to disclose their algorithm details, the publicly funded stuff is usually available for anyone who wants to take the time to leaf through PAMI or any number of other technical journals. Universities study this stuff with publication as a primary goal, so it's just a matter of knowing where to look. MIT's CBCL and CMU's Face Group are two of the better-known groups working on this kind of stuff, but there are others. Even if the researchers do not make their code available (and many do), it isn't too hard to put together an implementation and open source it yourself, as the algorithms themselves are publicly available in journals. I know because I implemented such an algorithm in a course last semester.


      The hard part is figuring out the little details that often get inexplicably omitted from journal papers. What are the particulars of the dataset? How are the training images preprocessed? What is the arbitration strategy for overlapping detections? These are the types of details that seperate the output quality of systems that use identical algorithms. In many cases, the researchers are happy to answer questions via email, unless they have plans to spin the research off into a private company.

    2. Re:open source/academic projects? by jon_galloway · · Score: 1

      I stumbled across this on SourceForge a while ago: http://cvs.sourceforge.net/viewcvs.py/cojojion/ They never released anything, but the code runs and it does a pretty good job. Uses genetic algorithms.

  56. A slight correction... by raygundan · · Score: 1

    I am mixing up a couple of different conditions in my previous post. (It has been eight years since I read that book last)

    I believe what the article was referring to was prosopagnosia. This condition has people able to tell a face from a vase, but unable to tell one face from another. The condition I was referring to I can't find the proper name for, but is more severe still, and may have been limited to the few cases for which the book I mentioned was named.

  57. I wonder what it.... by Anonymous Coward · · Score: 0

    I wonder if it would be able to detect the 'Red-Eye of Goatse'...

  58. USAF Research in the late '70s by cbelt3 · · Score: 1

    I worked on a project like this in 1979 as an undergrad research assistant at an engineering school in Ohio. It was an USAF project. The key to the research was that there were a discrete number of points on the human face (mostly relating to underlying bone structure), and the vectors between the points provided a unique and relatively unchangeable identifier. We used some (for the time) slick image processing algorithms to figure it all out- and had, as I recall, around 85% hit rate.
    Your defense dollars at work, kids.

    (Of course, I was in the project for the lab key- gave me a nice place to take girls on weekends..)

  59. Beer goggles? by NineteenSixtyNine · · Score: 0

    I have this problem with ex-girlfriends.

    --

    --
    What would Bill Clinton do?
  60. jesus in a pancake, jesus on a wall,... by Anonymous Coward · · Score: 0

    Jesus in a Lava lamp,
    I cannot count them all.

    Jesus was a prophet,
    His friends were all quite fine,
    But judging from their stories,
    They drank too much red wine.

    If Jesus'd come again today
    and multiply my paystub,
    I'd be very happy
    to pick up the dinner tab.

  61. Even humans get false positives by Ra5pu7in · · Score: 1

    We aren't computers, yet we sometimes look at the wall or the ceiling or the stars and pick out faces. Many people have scared themselves badly seeing a face in the dark woods or looking in a window at night.

    --
    I was taking one day at a time, but then several days got together and ambushed me. (from a Rhymes with Orange comic)
  62. Two Faced Bush by sYn+pHrEAk · · Score: 2, Funny
  63. bush and chimp by super-momo · · Score: 0, Troll

    http://vasc.ri.cmu.edu/demos/faceindex/05062004/us ers/2250.html

  64. How about with a terminator style HUD? by phorm · · Score: 4, Funny

    Target Identified:

    Name: Darl McBride
    Employer: SCO
    Karma: -5000

    Options: Disable target, bring to penguin HQ for further questioning
    Brute force: Authorized and encouraged.

  65. This is what it's like. by Anonymous Coward · · Score: 2, Informative
    I have prosopagnosia, and this is what it's like.

    Faces are moderately recognizable for me, but no more so than other objects, like rocks or cars. See this page, where someone else with this problem has a demonstration page.

    Prosopagnosia is rare. Only about thirty people in the US have been formally diagnosed. I have; there's a researcher at U.C. Berkeley who ran me through the tests. There's a specific section of the brain that does face recognition, and it goes active when looking at a face. This can be detected with a functional MRI scan, which takes hours and involves looking at pictures while inside an MRI machine. For people with prosopagnosia, that doesn't happen.

    It's a social handicap. It's most annoying in medium-sized groups. In big groups, you're not expected to know everyone; in small groups, cues other than faces are sufficient. It's subtle. One of the most subtle effects is that recognition takes well under a second for people who can recognize faces, but may take two or three seconds if you have to do it by other means. This breaks some implicit social cues.

    Recognizing people works about equally well from the front, side, and back for me. Voice and walk are more helpful than seeing the face.

    Practice doesn't help. The non-face recognition skills can be improved, but that's a workaround. Real face recognition, the kind that makes reading People magazine meaningful, is totally out of reach.

    "Falling in love" doesn't work, either. Sex, yes; friendship, yes, love, no. That's tied to face recognition. As a friend of mine puts it, "there's no click".

    It hasn't interfered with professional success for me; I have an advanced degree from a big-name school and I'm a multimillionaire. But it leads to a strange life.

  66. ROC by feelyoda · · Score: 1

    You really need to appreciate what is going on here...

    Face detection, as any detection task, requires that a class of object be identified. This doesn't mean we have a picture, and ask, "is it a face?". It means we have a 640x480 image, say, and 5 scales, and want to ask which of the ~1,500,000 possible locations is a face.

    To say that there are a few false detections means that this detector has a false positive rate of O(1)/O(10^6), which is awesome.

    There is a natural trade off between detection rate and false positive rate. You can imagine that just saying everything is a face yields a perfect detection rate, and a horrible false positive rate.

    Plotting detection rate vs false positive rate yields a receiver-operator-curve, which is really what you are looking for to compare face detectors. It would be hard to find a better one than what they have done at the Robotics Institute.

    --

    Robo-Blogs of the world: UNITE!
  67. dont forget by Anonymous Coward · · Score: 0

    the face on Mars

  68. I said that. by Positive+Charge · · Score: 1

    Rather, I once said (but got rejected) that I want a public face database. Presumably, your face is not subject to reasonable expectation of privacy, so people could compile face databases. I could use it to get the names and criminal records of my daughter's boyfriends, to help identify burglars in my home, to remember people I don't remember (with a little camera on my PDA?)

    We need a central repository of names/faces which anyone could openly download.

    THEN I will open a business selling IR reflective makeup to spoof the cameras, (because I'm a greedy bastard (Libertarian)).

    Heck, I could even use it to identify undercover cops investigating my open WiFi!

    Oops.

  69. Someday... by Anonymous Coward · · Score: 0

    it will also be able to detect Hanibal and B.A. Baracas too!

  70. Wow! by The+Spoonman · · Score: 2, Funny

    Once this thing works, I'll be able to Google for porn done by my favorite actresses, regardless of correct file names! YAAY!!

    --
    Which is more painful? Going to work or gouging your eye out with a spoon? Find out!
    http://www.workorspoon.com
    1. Re:Wow! by igny · · Score: 1

      What we really need now is to sort out mp3s on p2p: fake albums from real regardless of filenames.

      --
      In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
  71. Names to faces? by TXP · · Score: 2, Interesting

    My problem remembering people's names stems from the countless hours playing video games. In video games the characters don't have feelings so it doesn't matter if you remember their name or not. So I tend not to remember their names but what they function/occuption is. So if someone tells me they are George the CEO of the company I work at. I'll remember that he's the CEO of the company but not his name.

  72. Realtime Verification by nurb432 · · Score: 2, Insightful

    As you upload your images to your PC, it verifies you havent taken any pictures of 'banned' individuals... Such as movie stars...

    If you have, it deletes the image, and e-mails the MPAA.

    Though I'm joking, there is room for such abuse once you have to be authenticated even to view your own images.

    --
    ---- Booth was a patriot ----
  73. Ask an artist by JGski · · Score: 2, Interesting
    The funny thing is you can tell what matters for facial recognition by just asking or being an artist. Anyone who's done drawn or painted portraiture knows that it's the eyes, then mouth, and then nose that defines a recognizable face. The "proof" of concept is how you can draw a face with as few as 4 or 5 curved lines and the face is utterly recognizable. This is akin to Douglas Hofstadter's article "Letter Spirit" in "Fluid Concepts and Creative Analogies" which talks about what gives an image of a letter 'a', it all it's possible fonts and glyphs, its recognizable "letter-a-ness" or letter spirit as an 'a'.

    Most of the "insights" about facial recognition in the article would probably elicit a collective "well, duh, that's been known for hundreds of years" from artists (it does from me). BTW I identified all four faces correctly from the side-bar - being an artist let's you actually "see" the world rather "project" the world.

    JG

  74. Two eyes and one mouth by Anonymous Coward · · Score: 0

    Arrr ye insensitive clod, what do ye think this patch is for!? Yar!

  75. Monkeys compared by vuo · · Score: 0
    This submission struck me. In case you haven't seen it before, it's the famous collection of image pairs of Bush and a chimpanzee. True, the system can pick things that don't even remotely look like faces. But this image shows that we a still very, very far from a dependable system.

    All 20 Bushes are recognized correctly. The only anomaly is a repeated recognition of one face; the system sees two faces at the same location. But on the other hand, only one chimpanzee face of 20 is recognized in full and one by its eyes. In 15 of the 19 chimpanzees, the system sees no face at all. The remaining two chimpanzees show a false positive.

  76. Augmented by Uruviel · · Score: 2, Funny

    I'll have that feature augmented the next time I got surgery. "Who's that?" *program initiated* "That is your ex wife, reccomend 500meter distance from target" A nice female voice tells me.

  77. FYI by rms_nz · · Score: 1

    Just FYI on the topic of human face recognisition, this subject was covered in one part of the BBC's The Human Mind series...

  78. Re: Okay, call me crazy by Dog135 · · Score: 1

    Being autistic I have a similar problem.

    And yes, I often can't recognize my own family. My wife tells her friends online that if she wanted to hide from me, all she'd have to do is cut her hair. (whenever someone in my family changes their hair style, I really have to focus to see who they are)

    My sister, who I was liveing with at the time, helped me get a job at a place she was working at. When I showed up on the first day, the secretary came up to me and realy started talking up a storm! Well it was my sister. But she was wearing different clothes, hair style, and a glob of makeup. I don't think I ever did get it into my head that they're the same person. heh

    On the plus side, autism does make me VERY good with computers. They're so logical and understandable. And consistant if you're not running windows.

    --
    "That's so plausible, I can't believe it!" - Leela
  79. Re: Okay, call me crazy by The+FooMiester · · Score: 1

    All anyone in my family has to do is not talk around me. The only people who are recognizable to me are those with very distinct features. Such as my one ex-coworker who's overweight and walks with a very pronounced limp. Or my buddy Raven, who's 7' tall with long dark hair, and skinny as a rail.

    But as soon as someone opens their mouth and says something, I know them if I've talked to them before.

    --
    The previous has been a secret message to my comrades.
  80. Oblig. simpsons quote... by G-funk · · Score: 1

    ... Possible homer-sexual

    --
    Send lawyers, guns, and money!
  81. Red-eye in photo saved boys life by Daemonic · · Score: 1

    Don't go automatically editing the photos too enthusiastically. One mother detected cancer in her kid's eye when only one of them came out with red-eye in a photo.