Slashdot Mirror


Facebook's 'Rosetta' System Helps the Company Understand Text Within Image, Which is Crucial In Handling Memes, Flagging Abusing Content (techcrunch.com)

Facebook announced on Tuesday a new AI system, codenamed "Rosetta," which helps teams at the company as well as those at Instagram identify text within images to better understand what their subject is and more easily classify them for search or to flag abusive content. From a report: It's not all memes; the tool scans over a billion images and video frames daily across multiple languages in real time, according to a company blog post. Rosetta makes use of recent advances in optical character recognition (OCR) to first scan an image and detect text that is present, at which point the characters are placed inside a bounding box that is then analyzed by convolutional neural nets that try to recognize the characters and determine what's being communicated. This technology has been in practice for a while -- Facebook has been working with OCR since 2015 -- but implementing this across the company's vast networks provides a crazy degree of scale that motivated the company to develop some new strategies around character detection and recognition.

45 comments

  1. crucial at suppressing speech by Anonymous Coward · · Score: 3, Interesting

    It will be used to suppress speech.

    1. Re:crucial at suppressing speech by AmiMoJo · · Score: 1

      AC, what is your solution?

      --
      const int one = 65536; (Silvermoon, Texture.cs)
      SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
  2. Abusive Content? by Jarwulf · · Score: 2

    How is content ie simple information/knowledge abusive? Does it come out of the screen and berate you?

    1. Re:Abusive Content? by elrous0 · · Score: 1

      Haven't you heard? Words are now considered violence.

      https://www.nationalreview.com...

      --
      SJW: Someone who has run out of real oppression, and has to fake it.
    2. Re:Abusive Content? by Anonymous Coward · · Score: 2, Interesting

      Think back to the very worst moment of your life.

      Now imagine that somebody recorded a video of it and posted it to Facebook.

      To me, that video is "simple information".
      To you, the video is not just "simple information".

      Should the video be taken down?
      Let's consult the Platinum Rule: Treat others the way they want to be treated
      My answer is your answer to the question: Do you want the video of the worst moment of your life taken down?

    3. Re:Abusive Content? by butchersong · · Score: 2

      No. That seems very silly to me. What if the worst moment of my life was the towers coming down in NY? Does that mean that no one gets to post videos of that day? I may be able to get behind owning images of myself and taking down videos of me personally (though I doubt it) but I don't get to own and suppress ideas.

    4. Re:Abusive Content? by Anonymous Coward · · Score: 0

      Wow, way to wander off tangent. If you didn't think this meant a personal moment of some kind:

      Do you want the video of the worst moment of your life taken down?

      Are you even capable of having an intelligent discussion with other people?

    5. Re:Abusive Content? by Jarwulf · · Score: 1

      The vast majority of 'abusive' content is not that. The OP implies they are deleting speech. The majority of 'abusive' video examples are amateur porn girls freely distributed to one or more people then suddenly decided they wanted to take back. The very same people that want to impose 'right to be forgotten porno edition' as a felony matter also say that porno and the human body is no big deal and liberating. So logically circulating amateur porn is no different from anything else and if amateur porn has a right to be forgotten then someone like Trump can demand that anything embarrassing about him has a right to be forgotten.

    6. Re:Abusive Content? by Calydor · · Score: 0

      To follow your tangent, do you want the video of your kid dying as the tower collapses taken down?

      --
      -=This sig has nothing to do with my comment. Move along now=-
    7. Re:Abusive Content? by Anonymous Coward · · Score: 0

      Treat others the way they want to be treated
       
      And if I were a dirty Trump faggot I'd want to be called out for being a little shit eating fag.

    8. Re:Abusive Content? by Jarwulf · · Score: 0

      sorry should have added 'along the lines of what you describe'. 'Abusive' video content is more likely to be stuff like anyone proclaiming a conservative opinion.

    9. Re:Abusive Content? by butchersong · · Score: 2

      The AC didn't specify that the video had to include an image of me personally or one of my loved ones. That is a narrower argument but doesn't address the story posted for Facebook. We're talking about "hurtful" content. I may feel some personal attachment to a certain idea. Say my parents died in a camp in WW2 or something of that sort. That doesn't mean I should be able to suppress holocaust denial. Say instead that someone was posting pictures of my dead parents in the camp, in that case I can see an argument but I'm not sure still which side of it I would end on. In this case we're not even talking about images themselves but ideas communicated in the images through text. So I don't think my comment the one going off on a tangent. I think the AC's comment was the tangent... unless I'm misunderstanding the story.

    10. Re:Abusive Content? by Anonymous Coward · · Score: 0

      With a 6-digit user ID, don't you remember Goatse or Tubgirl on Slashdot. That's how. We were born into internet abuse, molded by it. We did not see a code of conduct until we were already men; it was limiting.

    11. Re: Abusive Content? by Anonymous Coward · · Score: 0

      We want people who type things like that to not be suppressed. You should be free, and not feel afraid to have your real name tagged alongside. Because you shouldn't be prohibited from expressing yourself and experiencing the consequences thereof.

    12. Re:Abusive Content? by Anonymous Coward · · Score: 0

      Wow, you are an asshole.

    13. Re: Abusive Content? by Anonymous Coward · · Score: 0

      Because we can't put you on some "shitlist" or blocklist without your real name and a registered account. Cough it up faggot, you scared of being blackballed everywhere online if you associate your identity with your email address? Scared of being judged by your probably lengthy online past history before you even open your mouth?

    14. Re: Abusive Content? by Anonymous Coward · · Score: 0

      Down with freedom of speech! Heil Hillary!

    15. Re:Abusive Content? by Anonymous Coward · · Score: 0

      Wow, you missed his first sentence. It read:

      No.

      Are you even capable of removing your blinders?

  3. Translation by Anonymous Coward · · Score: 3, Insightful

    "We can now identify conservative and Trump-supporting users much more rapidly in order to ban them for #WrongThink!"

    1. Re:Translation by Anonymous Coward · · Score: 0

      Political affiliation is just a coincidence.

      The real reason is that a bunch of fat cunts, faggy jews and ugly minorities don't want unflattering realities rubbed in their disgusting faces. Because the words are true, and truth hurts. A "safe space" is where they can be ugly and unwanted and snivel about their stupid feelings without being humiliated in response.

    2. Re: Translation by Anonymous Coward · · Score: 0

      Pretty much this. Trump supporters meme better because unlike leftists and feminists we have a sense of humor. So of course memes have to be censored.

    3. Re:Translation by Anonymous Coward · · Score: 1

      fat cunts, faggy jews and ugly minorities

      Which job you want? you want to fire up the ovens or shovel coals into the furnace?

  4. LOL. Just license the Battlefield V chatbot by elrous0 · · Score: 1, Interesting

    You know, the one that labeled "white man" as hate speech.

    https://www.forbes.com/sites/e...

    --
    SJW: Someone who has run out of real oppression, and has to fake it.
  5. Ray Ban? by aitikin · · Score: 1

    Yet I always see the same Ray-Ban "Sale" spam image/post from hacked accounts...you'd think they'd train it to remove that...

    --
    "Don't meddle in the affairs of a patent dragon, for thou art tasty and good with ketchup." ~ohcrapitssteve
    1. Re:Ray Ban? by Anonymous Coward · · Score: 0

      Especially since it's an unpaid ad. Well, probably.

    2. Re:Ray Ban? by Anonymous Coward · · Score: 0

      Or the Honda ads. Honda has the lowest IQ owners so of course they're the chosen target of scammers. It's a self-selecting group of stupid.

  6. Good to know by SuperKendall · · Score: 1

    Pretty easy to include some nonsense words that would get parsed by the AI but not by the viewer...

    --
    "There is more worth loving than we have strength to love." - Brian Jay Stanley
    1. Re:Good to know by Anonymous Coward · · Score: 0

      I'd start with some radical fonts, or maybe even cursive. Sort of like those stripes you put on your face to fool facial recognition algorithms.

  7. Used to filter ads by Dan+East · · Score: 3, Insightful

    Of course this tech is being spun to "save the children", but it is also used to screen all advertisements that run on FB. They do not want ads to contain much text - less than 20% of the area of the ad image can be text. This is detected automatically using the technology described, and their system will stop the ad if it doesn't meet that requirement.

    We've found that images with less than 20% text perform better.
    To create a better experience for audiences and advertisers, ads that run on Facebook, Instagram and Audience Network are subject to a review process that looks at the amount of image text used in your ad. Based on this review, ads with higher amounts of image text may not be shown. Keep in mind that some ad images may qualify for an exception. For example, book covers, album covers and product images usually qualify for an exception.

    https://www.facebook.com/busin...

    And from the blurb:

    detect text that is present, at which point the characters are placed inside a bounding box

    Thus the area of the bounding boxes (after performing a union) can be at most 20% of the area of the image.

    --
    Better known as 318230.
  8. Revenge time has come! by devslash0 · · Score: 3

    Now we should start writing text on memes using Captcha fonts.

    1. Re:Revenge time has come! by Mal-2 · · Score: 1

      If that's what it takes to kill off Comic Sans for good, then let's get to it.

      --
      How is the Riemann zeta function like Trump rallies? Both have an endless number of trivial zeros.
    2. Re:Revenge time has come! by Anonymous Coward · · Score: 0

      I welcome Facebook's new CAPTCHA-solving overlords, aka. people will just get more inventive and use ASCII art or something or confuse its OCR capabilities.

    3. Re:Revenge time has come! by Anonymous Coward · · Score: 0

      Came here to say the same thing. mod parent up.

    4. Re:Revenge time has come! by Mal-2 · · Score: 1

      Or just spell things in 1337 or like "a møøse once bit my sister". These are still detectable, but should drift from one meme image to the next, which means that instead of one large group, they'll fall into many small groups, which should make them harder to associate. "Fly under the radar" as it were.

      --
      How is the Riemann zeta function like Trump rallies? Both have an endless number of trivial zeros.
  9. imperial delusions, an empire in the sand by harvey+the+nerd · · Score: 1

    That sociopath/psychopath MZ fancies himself as "Augustus", the first Roman emperor, but neglects which Augustus really corresponds better.

    Hopefully, Romulus Augustus, aka Momyllus Augustulus, might be better for the position on the timeline of MZ's empire...
    One of the other Augustus over several hundred years probably better reflects his pathology...

  10. Fuck! by Anonymous Coward · · Score: 0

    Fuck Zuck! That penis-weiding castration of a very short, short man. What a disgrace you must feel when your wife has to get off her knees to approach your wilty pod.

  11. Capcha by duke_cheetah2003 · · Score: 1

    I imagine now we'll see internet meme in those awful fonts with twisted up letters that we humans can figure out in microseconds, but a computer is completely baffled.

  12. the chosen ones have spoken and decided by Anonymous Coward · · Score: 0

    The regulars arent allowed to hear or see whatever they'd like.

  13. Use only Comic Sans by jfdavis668 · · Score: 1

    They will never be able to detect that.

  14. Wrongthink detector by Anonymous Coward · · Score: 0

    Call it what it is.

  15. Abusive content? by PPH · · Score: 1

    So they've come up with a way to identify that silhouette of the guy with the big nose?

    --
    Have gnu, will travel.
  16. So no more by AHuxley · · Score: 1

    Memes? Art? Cartoons? Blasphemy?
    No more freedom of speech. No right to assemble art work and publish?

    --
    Domestic spying is now "Benign Information Gathering"
  17. Captchas by Anonymous Coward · · Score: 0

    Now let's just hope it can be used to solve Captchas.

  18. CAPTCHA Memes... by Anonymous Coward · · Score: 0

    will now be the new thing on Facebook. AI vs AI, who will win?

  19. Sanction Israel? by Anonymous Coward · · Score: 0

    So Trump is going to sanction Israel because of the interference of AIPAC, which "vastly overwhelms" anything Russia has done?

    https://www.independent.co.uk/news/world/americas/us-politics/israel-us-elections-intervention-russia-noam-chomsky-donald-trump-a8470481.html