Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror
×
AI Science

Can Machine Learning Guess True Emotions From Facial Microexpressions? (cmu.edu) 55

jbmartin6 writes: Microexpressions are fast, involuntary facial expressions which other people may not consciously recognize, but arise from our real emotions instead of the face we wish to present to the world. Carnegie Mellon University released an interesting blog entry about new approaches to using computers to recognize these microexpressions with a focus on the security and military applications. If you haven't taped over the cameras on your devices, it might be time to start thinking about it. Just imagine how advertisers would (mis)use this sort of technology.
"Our approach uses machine learning features that treat the whole face as a canvas," writes the lead researcher, adding "One challenge we faced for this project was finding a dataset with accurately labeled data to establish ground truth.

"Few existing databases capture subjects' suppressed reactions...."
This discussion has been archived. No new comments can be posted.

Can Machine Learning Guess True Emotions From Facial Microexpressions?

Comments Filter:
  • Sounds like lie detectors, with a huge potential for false positives.

    • I would be very interested if A.I. Would answer the god dawn question. I get enough liars just watching commercials on TV
    • you get a false positive out of it and that lets you get the warrant & probable cause to go on the fishing expedition you wanted to.
    • "Sounds like lie detectors, with a huge potential for false positives."

      Easily beaten with a few Botox injections.

  • And the problem starts with establishing ground-truth: Those that are into deception will not give you a sample or will taint the data-set intentionally. From the press, I gather that the average human adult lies once every 10 minutes. Sounds too high to me, but still, almost everybody lies regularly and hence almost everybody has reason to taint the basis for ground-truth.

    But the fascist fuckers that want people to be completely unable to have secrets will keep trying...

    • by Actually, I do RTFA ( 1058596 ) on Sunday January 21, 2018 @01:59PM (#55972541)

      the average human adult lies once every 10 minutes. Sounds too high to me

      Sounds high to me too. Now, how I am going to kill 9 minutes and 59 seconds.

    • by Nutria ( 679911 )

      From the press, I gather

      The same press that regularly hires "reporters" who commit plagiarism, make up stories from whole cloth to fit their political biases, use "confidential sources" which have their own obvious agendas, and accuse people of rape without a hint of evidence because, you know, All Men Rape?

      That press?

    • From the press, I gather that the average human adult lies once every 10 minutes.

      "The press" reached this conclusion by observing behaviour around their own offices. Unfortunately their approach was deeply flawed since they assumed that press-scum was representative of "the average human".

  • by RyanFenton ( 230700 ) on Sunday January 21, 2018 @01:59PM (#55972543)

    Here's the thing about emotions. They don't mean what you'd expect. Emotions ARE logic to an extent - short-circuited logic, quicker and made with incomplete information. They're the quick-and-dirty half-logic that makes us able to function in a world of unknowns for millions of years as a species.

    They aren't 'serious business' - they're formed and dismissed in fractions of a second, and most of them don't trigger facial emotions - and even then a raised eyebrow muscle can mean confusion, pronounced indifference, mild surprise, or just positioning to move your eyeball, or many other things.

    Like most things, this ain't mindreading - it's just more polygraph logic being applied, ironically like emotion itself is applied - for the basis of 'asperational truth', or bullshit they hope might pan out.

    Ryan Fenton

    • by Greyfox ( 87712 )
      Yeah. Often when playing poker, you can tell that your opponent is getting a huge adrenaline rush. But is he getting a huge adrenaline rush because he's bluffing and hopes you'll fold, or because he's sitting on a huge hand and hopes you'll call?
  • Well, even assuming that micro-expressions can indeed reflect some mental state, how would you teach an AI that ? It's not like you can have 10000 pictures of people with a valid set of associated micro-expressions... simply because nobody can accurately recognize micro-expressions except in some edge cases. They are very dynamic too, so a picture won't cut it (a video might).

    And about the taping of video cameras on PCs, why aren't the LEDs hardwired to the camera: if the camera is ON, the LED is too, no
    • by AHuxley ( 892839 )
      Why are so many social media sites asking for a baseline profile pic for "reasons" as a baseline.
      The AI knows what a normal user looks like. The AI can be educated by SJW about what past comments got banned and accounts removed.
      Connect up the past images of the banned person as they created the text of the wrong type of comment and thats the SJW AI learning.
      Use that created data set to detect other real time users trying to type messages that SJW want banned.

      The detection of a problematic users face
  • No ,,, (Score:4, Funny)

    by CaptainDork ( 3678879 ) on Sunday January 21, 2018 @03:38PM (#55973105)

    ... in some cases, it will return an answer of, "pissed off gorilla."

  • by tgrigsby ( 164308 ) on Sunday January 21, 2018 @03:53PM (#55973201) Homepage Journal

    Judging from what I saw at NRF in New York City earlier this week the answer is a definite no. Not only did the systems I saw fail to guess my facial emotion but they also decided that I was female. I'm a 53 your old man with short hair and a goatee. It kind of makes you wonder what kind of faces they used to program in what a female face would look like...

    I don't think that we are far off but that conference tends to have some pretty decent cutting-edge systems to show off and we are clearly not there yet.

    • Re:Not yet (Score:4, Interesting)

      by goose-incarnated ( 1145029 ) on Sunday January 21, 2018 @05:56PM (#55973965) Journal

      Judging from what I saw at NRF in New York City earlier this week the answer is a definite no. Not only did the systems I saw fail to guess my facial emotion but they also decided that I was female. I'm a 53 your old man with short hair and a goatee. It kind of makes you wonder what kind of faces they used to program in what a female face would look like...

      I don't think that we are far off but that conference tends to have some pretty decent cutting-edge systems to show off and we are clearly not there yet.

      I started getting into ANN and ML recently, just as a hobby. My finding thus far is that the AI we are hearing so much about is nothing more than a sophisticated fuzzy pattern matcher. That's literally all it does: it "learns" by mapping patterns to a limited range of predetermined outputs. The path taken to determine what input should be mapped to what output is reinforced by feeding the machine millions of inputs (and providing the correct output to adjust the weights taken by the path).

      It's sophisticated, computationally expensive, and hilariously nowhere near "intelligence". After seeing how these machines (the ANN) functions I'm pretty certain that we're nowhere close enough to intelligent machines. Sure, we can train a particular NN to beat humans at Go, but the first time we put it in front of another game (recognising pictures) it will lose. Give it enough time and error-correction and it will eventually win that too, but the problem is that there are so many "general" tasks that humans and other animals accomplish with much less training that the machine can never hope to keep up.

    • by q4Fry ( 1322209 )

      ...also decided that I was female. I'm a 53 year old man with short hair and a goatee. It kind of makes you wonder what kind of faces they used to program in what a female face would look like...

      Ah, but what species did it suggest you are?

  • Guess accurately? That could be interesting...

    How cool would it be to literally have an emoji be part of your password, where the emoji is your literal face, and not some predefined unicode value?

  • by maiden_taiwan ( 516943 ) on Sunday January 21, 2018 @09:58PM (#55975157)

    >Microexpressions are fast, involuntary facial expressions which other people may not
    >consciously recognize, but arise from our real emotions instead of the face we wish to
    >present to the world.

    FYI, that is only a hypothesis, not a universal truth, and there's plenty of evidence against (and debate about) so-called microexpressions and "unconscious emotion." For example, if you place electrodes on people's faces and measure actual muscle movements during emotion, there's tremendous variety, not uniformity. There's also lots of evidence that emotional expressions in the face are neither universal nor innate [affective-science.org].

    See the recent TED Talk by neuroscientist Lisa Feldman Barrett [ted.com] for a friendly overview.

  • Whatever the merits of the situation I see a business opportunity for people who make protective cases for cell phones. Simply build one with an easy to install and easy to pop open attached lens cap. "Let 'em eat darkness."

    {^_-}

Never buy what you do not want because it is cheap; it will be dear to you. -- Thomas Jefferson

Working...