Stories
Slash Boxes
Comments
Search

News for nerds, stuff that matters

Slashdot Log In

Log In

Create Account  |  Retrieve Password

Virtual Robots Fooled By Visual Illusions

Posted by kdawson on Sun Sep 30, 2007 05:45 PM
from the not-as-they-seem dept.
Robotics Science
Roland Piquepaille alerts us to research out of University College London in which virtual robots, trained to "see" as we do, were duped by optical illusions the same way humans are. Here's one of the illusions the software system fell for.
+ -
story

Related Stories

Submission: Virtual robots fooled by visual illusions by rpiquepa (644694)
This discussion has been archived. No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.

Virtual Robots Fooled By Visual Illusions 50 Comments More /

 Full
 Abbreviated
 Hidden
More
Loading... please wait.

  • That's not an optical illusion (Score:3, Funny)

    by nurb432 (527695) on Sunday September 30 2007, @05:49PM (#20804463) Homepage Journal
    That is just a difference in lighting.

    • Re:That's not an optical illusion (Score:5, Informative)

      by cathector (972646) on Sunday September 30 2007, @05:59PM (#20804539)
      no, it's not a difference in lighting.
      the central squares are in fact the same color on your monitor, (pretty close to hex: 647316).

      this is very similar to this famous color constancy illusion [wikipedia.org].

      • Re: (Score:3, Informative)

        by alexhs (877055)
        I think that's not what the parent meant.

        Lightning change through the day, so the actual color of reflecting objects also is changing. But the object didn't physically change and your brain "corrects" color, that is abstracts them (you wouldn't say your blue car to be blue the day and dark gray the night, it's simply blue).

        In the illusion at hand, left sphere is interpreted as being lit by a red light, while the red sphere is interpreted as being lit by a blue light.

        Of course, "Ceci ne sont pas des sph

        • Re:That's not an optical illusion (Score:5, Informative)

          by nine-times (778537) <nine.times@gmail.com> on Sunday September 30 2007, @07:03PM (#20804873) Homepage

          Of course, "Ceci ne sont pas des sphères", only pixels, so the comparison on interpreted colors fails.

          That's what makes it an optical illusion. Your brain is interpreting visual information based on a context which causes a failed interpretation. That could be a definition for "optical illusion".

          These aren't colored spheres, and no one said they were colored spheres. It's just an arrangement of colored patches, arranged in such a way as to give your mind a bunch of visual cues that there are different colored lights shining on those patches, causing your brain to misjudge the actual color of those colored patches. Hence, it is an illusion.

          • Ya my brain had failed interpretation too ... and ... *sniff* *sniff* ... does anybody else smell burnt toast?

        • Re: (Score:3, Informative)

          by Tank720 (1164591)
          It's not about lighting at all. It's about lateral inhibition. The neurons in your brain, when connected in a nural network, send inhibition signals to their neighboring neurons. Areas close on you retinas have neurons on the same part of the brain, so the squares surrounding the center square are going to activate neurons close to the ones you use to see the center square. Now the lighter squares send a stronger signal to the neurons surrounding the center square, thus sending MORE inhibition, thus ma
        • not an error on the part of the computer vision algorithm.

          When I first looked at the illusion I didn't see the spheres so much as one in red light and one in blue light, but though they were two different plaid buttons. The "trick" didn't work so well for me. I went back and looked at the illusion again and this time noticed the "puddles of light" at the bottom of each sphere and my visual cortex reinterpretated and the "trick" worked.(try looking at the illusion with the puddles of light covered up and t

        • That's not the same thing. When you put a white tile under a red light vs a blue light, then photograph it, the sensor (and our eyes) are going to register the red and blue light being reflected from the tile and it really will show up as red or blue (well, YMMV on the actual shade detected).

          In this case, the center squares in the image are the same exact color and it has nothing to do with lighting. The pixels don't lie. Our eyes (being tied to the crazy pattern-recognition system our visual cortexes

    • More exactly : diffent surrounding... (Score:5, Informative)

      by DrYak (748999) on Sunday September 30 2007, @06:58PM (#20804843) Homepage
      ...that's a difference in surrounding lightning.

      Human visual system (as most other senses) work not by absolute values (i.e.: it doesn't see that the color '#c0ff20' or whatever), but mainly by comparing the signal with signals from the surrounds.
      Thus what we technically see is that on the left object the central case looks much more "greener" than its surrounding, in the right object, the central case is "much more orange" than the surrounding. In fact, when the mask is enable, the colours do change from the point of view of the visual system : we were seeing contrast with two different surrounding, now we see a contrast with a third surround (mostly black). We see three different contrasts, even if from the computer's point of view the color is them same (the same RGB triplet / same intensity on your CRT/LCD)

      If the scientist are trying to build efficient visual systems, they are probably mimicking this "works-by-comparing" method that the nature is using.
      That's why we can recognise the same object, during day, during night, with weird lights, displayed on the screen (worse colour gamut) or on a print out (even worse color range). Because the relative difference stay the same, even if the colour as-seen-by-a-computer change.

      The same is valid for any other sens, or in fact, any other information that is processed by neurons. Everything works by comparing (across several signals, across time, etc.). There's no such thing as "an absolute value" in the information carried by neurons.

      That's also why all those "but the human eye can only x thousands of colors" (usually mocking the latest 32bit, 48bit, floating point or whatever color depth), are fundamentally wrong.
      Yes, the human visual system can only distinguish a hundred or so colors.... ...WHEN those colors are ISOLATED. (i.e: putting the name "red" "orange" "purple" on a color you see alone).
      When two colors are put next to each other, the human brain can suddenly distinguish much more subtle variations (each color would be considered as "brown" when seen alone, but next to each other, you can use thousands of different shade of brown and the eye will still see the difference).

      That's also why radiologist are fond of high contrast / big depth screens : because all those difference in shades of grey *can* be distinguished and *are* revellent for the diagnosis when displaying X-Ray pictures.
        • The illusion is constructed to look like the different colors of the surrounding area are caused by different lighting (the object looks the same, except for the color, the color difference is such that it can be caused by tinted light and there is an indication of shadow and surrounding tinted light on the "ground"). Our visual system would still be fooled if they had just split a 2D checkerboard in half and colored it differently, but then we would be more cautious about our conscious assessment. The illu

  • Hoax! (Score:5, Funny)

    by ynososiduts (1064782) on Sunday September 30 2007, @05:50PM (#20804467)
    They could just be programmed to look as if they were falling for it! I smell conspiracy!

    • Or, maybe, they're faking it. So that we don't know how advanced they're getting, and wont see it coming when the robot revolution comes.

      So, let me go on the record now, saying: I welcome our soon-to-be-evolving robot overlords!

  • Colors sure look different... (Score:5, Interesting)

    by Gertlex (722812) on Sunday September 30 2007, @05:54PM (#20804509)
    ... but I went ahead and verified with a pixel color id program (ColorPix) that they are the same color.
    • but I went ahead and verified with a pixel color id program (ColorPix) that they are the same color

      Indeed they are, but for me at least, this illusion didn't seem as "abrupt" as others do when it's shown that the perception is false. One that always stands out to me is this one (many have probably seen it):

      Without thinking too much, look at the colors of the A and B squares [imageshack.us] in this well-known image.
      Now, here's an animation I just made showing the truth [imageshack.us]. That's a solid, unchanging color going from A to B.

      I think this a much more drastic difference than the one in TFS, but of course YMMV :)
      • What I've always admired about that particular illusion is that, even though you think e.g. "it's the shadow that's making them look different!", you can keep deleting things from the image that you think are causing the illusion, all the way down to the point where only those two squares are left, and they will still look different the whole time.

        Maybe make an animation demonstrating that? ;-)
  • to being able to upset them with goatse. Maybe that is what starts the robot uprising...

  • Model of Reality (Score:4, Insightful)

    by SpectreBlofeld (886224) on Sunday September 30 2007, @05:59PM (#20804537)
    From TFA: "The virtual robots in this study were driven solely by the statistics of their training history and used these statistics as the basis of their correct and subsequent incorrect decisions. Similarly, we believe the human brain generates perceptions of the world in the same way, by encoding the statistical relationships between images and scenes in our past visual experience and uses this as the basis for behaving usefully and consistently towards the sources of visual images." So the robot vision was created as a model of human vision, and it succeeded at doing so. That's sort of interesting, I suppose, but what does it tell us? That we were right about the way human vision works? Seems to me that the point here is really that in some ways, human vision is 'broken' and that maybe it isn't the best apparatus for machines to use. If we want to welcome our robotic overlords, we should be improving on the vision model, not trying to give machines the same flawed framework.

  • Welcome? (Score:5, Funny)

    by CriminalNerd (882826) on Sunday September 30 2007, @06:01PM (#20804553)
    I'm not sure about you guys, but at the moment, I'm kind of doubtful in welcoming our new robotic overlords. I mean, I thought they were supposed to be superior to us and not be fool by petty illusions...

    • Re:Welcome? (Score:5, Insightful)

      by Gonzoisme (1023685) on Sunday September 30 2007, @06:18PM (#20804643)
      I think it is important to problem small weaknesses into our robots. You know, just in case.
      • OK, +1 Funny, I understand.
        But, +1 Insightful ? Are you people really that scared of an autonomously intelligent being you have no control over? Look around you they're everywhere! (OK, we could plausibly argue over whether the beings around you currently qualify as intelligent, nevertheless I think you get my point)
      • Definitely. That's why every time I design a Kill-bot, I make sure that it has a pre-programmed kill limit.

  • Its called white balance.... (Score:3, Insightful)

    by AgNO3 (878843) on Sunday September 30 2007, @06:06PM (#20804583) Homepage
    and your eyes do it to. Which is why even sodium vapor lights don't look as yellow as they really are by the human eye. Turn off the white balance on the robots and I bet you they will see them as the same color. Add the average inverse color as a background for each color and your eyes will see them totally different. IE blue behind the orange and orange behind the blue. really stupid test.
    • But you don't want them to see the same color, because in the computer vision application it is important to deduce the object color. You can't just measure the reflected light. The optical illusion shows that, in our attempt to find the object color, we "guess" the light color from the colors of the surrounding scene and compensate for it, and so do the robots with automatic white balance. We're still fooled by light sources which aren't approximately black body radiators and robots face the same problems
    • Really more an artifact of retinal fatigue, I'd say.

  • Make them explode (Score:5, Funny)

    by Bender Unit 22 (216955) on Sunday September 30 2007, @06:10PM (#20804605) Journal
    show them a Escher staircase.
    • Good robotic overlords like the Daleks will never be stopped by a staircase, Escheresque or not - they simply demolish the building.

      Demolish! Exterminate! Exterminate!

  • Now try this: (Score:5, Interesting)

    by Wilson_6500 (896824) on Sunday September 30 2007, @06:13PM (#20804625)
    Cross your eyes, line up the two squares so they're offset by a few millimeters, and then hit the mask. What I saw was that the squares retained their seeming discoloration--until I uncrossed my eyes.
    • Actually, even if you hit the mask first, and only then cross your eyes, they still appear different. At least they did for me. Not sure what it means though.
  • This is why I turn my auto-white-balance off in my digital camera. If I need to adjust the color, I'll do it later in Photoshop. (Another reason to shoot in RAW mode.) -- Carey

  • The colors in the illusion look the same... (Score:5, Funny)

    by Junta (36770) on Sunday September 30 2007, @06:42PM (#20804747)
    But then again, being color blind makes a lot of things look the same that shouldn't be...

    • Re: (Score:3, Insightful)

      by pcgabe (712924)
      I came here to post the exact same experience. I'm red-green colorblind; they look the same to me.

      OT question, since you're also colorblind and I'm curious: does your girlfriend wear makeup? See, mine does.

      WHO IS SHE WEARING IT FOR?
      • >> WHO IS SHE WEARING IT FOR?

        Presumably herself.

        Slashdot lameness filter bypass:
        Reason: Don't use so many caps. It's like YELLING.
        Reason: Don't use so many caps. It's like YELLING.
        Reason: Don't use so many caps. It's like YELLING.
        Reason: Don't use so many caps. It's like YELLING.
      • The moment I saw the image - and before reading the text - I noticed that the right-side square looked to be a strange hue compared to the rest of the sphere. It stuck out to me because it looked out of place - more orange in it, amongst the bluer greens. When I visually compared it to the left side I saw why - it's the same colour as the central square on the left side! Note that this is how I *perceived* it initially. They didn't look different to me.

        Maybe my past experience at seeing such illusions has t
  • they are also susceptible to the illusion of "beer goggles"? ...Next thing you know, your personal robot's software has it waking up in bed with your new Dyson vacuum and a strange Toaster! [There must be a Bender joke in here somewhere] http://en.wikipedia.org/wiki/Bender_(Futurama) [wikipedia.org]
  • Since we are not born with flashlights glued to our heads (although in Soviet Russia one can be obtained by getting into a fistfight), we must compensate for the tone of ambient lighting. This correction that we easily do in our heads but must be applied manually on digital cameras in fact allows us to determine true color of the objects more accurately in natural settings. Therefore I wouldn't call this an optical illusion any more than the fact that our eyes become more sensitive at night.

  • Can someone explain what this teaches us? (Score:3, Interesting)

    by gozu (541069) on Sunday September 30 2007, @07:06PM (#20804889) Journal
    I'm struggling to find the utility of the study. So, if we learned to see differently, we could see the world in a way different enough to not be fooled by certain optical illusions, and probably be fooled by others?

    Assuming it is possible to change the way a human sees without breaking the brain. A popular theory on evolution is that we evolved our brains to better analyze visual data coming in. We're not deceived as easily by certain camouflages animals use. Stripes, dots, color, etc.

    Confirms what we thought about the way we learn to see, perhaps? That'd make sense.

    • Re:Can someone explain what this teaches us? (Score:5, Insightful)

      by David_Shultz (750615) on Sunday September 30 2007, @10:38PM (#20806221)
      I'm struggling to find the utility of the study. So, if we learned to see differently, we could see the world in a way different enough to not be fooled by certain optical illusions, and probably be fooled by others?

      The program wasn't designed to detect optical illusions -it was a by-product of the training the system went through. The fact that it was tricked by a similar illusion without being programmed to do so might be taken as suggestive that our learning mechanisms are similar to the ones used by the program. From TFA:

      The virtual robots in this study were driven solely by the statistics of their training history and used these statistics as the basis of their correct and subsequent incorrect decisions. Similarly, we believe the human brain generates perceptions of the world in the same way, by encoding the statistical relationships between images and scenes in our past visual experience and uses this as the basis for behaving usefully and consistently towards the sources of visual images
  • I'm normally affected by optical illusions just like everyone, but I don't see it here - the center squares of both discs appear equally light green to me. Any ideas? I don't think I'm colorblind...
      • But I don't see the illusion even with either one (and, obviously, with both ;-)) of my eyes closed.
  • If Virtual Robots are fooled by Visual Illusions, what does that tell us about real robots?

    In the virtual world I created, Smorgons are 6 meters tall, shoot acid out of their noses, and have been known to breed 10 offspring in a month.

    In the real work, however Smorgons don't exist, so therefore I must conclude that virtual tells us nothing about actual.

  • Ok, so what if your robot is fooled by some obscure optical illusions-- other 'illusions' (or for lack of a better term, optical phenomena) are far more problematic- consider the problem of recognizing that you're looking at a mirror and not just a big room. Or the problem of 'seeing' the subtle reflections cast by a transparent medium like a window, in order to recognize the presence of an obstacle. Speaking as someone who's done a fair amount of work on autonomous robot exploration, these are big unsolve

    • Ban Roland
      I've seen this sentiment on numerous occasions, but I don't quite understand it. What is wrong about him? I can't say that I can see anything obviously suspicious about this story.

        • Re: Ban Roland (Score:5, Informative)

          by The_Wilschon (782534) on Sunday September 30 2007, @09:12PM (#20805765) Homepage
          The general explanation, as I have seen it given many many times previously, is that, rather than write a /. story which links to some science/tech article, roland will paraphrase the article in his blog, and link the /. story to his paraphrase. This is a means of gaining ad revenue for himself and his employers (ZDnet, I think?), but it doesn't give any ad revenue to those who actually did and wrote up the research.

          Is this true? I don't know. I never RTFA.

          • Re: Ban Roland (Score:5, Funny)

            by shellbeach (610559) on Sunday September 30 2007, @09:21PM (#20805807)

            The general explanation, as I have seen it given many many times previously, is that, rather than write a /. story which links to some science/tech article, roland will paraphrase the article in his blog, and link the /. story to his paraphrase. This is a means of gaining ad revenue for himself and his employers (ZDnet, I think?), but it doesn't give any ad revenue to those who actually did and wrote up the research.

            Is this true? I don't know. I never RTFA.
            The links are to the original story. The "Roland Piquepaille" link goes to his blog, but it's unlikely anyone will be clicking on that one unless they're interested in the guy ... in which case, good for him.

            Incidentally, if there are any ads generating revenue on that blog, I'm not seeing them thanks to adblock. I doubt /. is the best place to try to get ad revenue, somehow ...

            • Re: (Score:3, Insightful)

              by Mr Z (6791)

              These days, Roland's links go to the original story. Originally, his links went to his blog, as others have described. I don't his current posts. His previous ones, though, were just a bit self-serving.

          • He USED TO link to his blog as the article links in his text, which had ads. Now he links directly to the articles he talks about in the /. summary, and his name's link is no_follow (I think).

            Pretty sure the Roland hate is from what used to happen and not anything he does now (AFAIK).
    • Whatever it was. I've pretty much thrown up my hands at trying to say something 'looks' green or orange or anything if it is in the realm I can't discern, because I recognize I simply have no comparative base to describe what something looks like to me in order to match a normal color vision person's perception.

      However, I do feel confident that it's accurately characterized in my case as a dramatic insensitivity to red, so it is a decent bet that to me that everything looks less red (i.e. brown looks green
    • I doubt we ourselves can make things that match or exceed our own capacity because then we would need to fully comprehend our own capacities which currently, we are far away from that (I work with brain research btw) AND replicate them in a correct matter.

      Now say we DO hypothetically develop something (like this vision stuff) that exceeds our own capabilities then in the very early beginnings of research and testing of it we would look at the output it creates, see extra data that isn't there from our viewp

All trademarks and copyrights on this page are owned by their respective owners. Comments are owned by the Poster. The Rest © 1997-2009 Geeknet, Inc.