Checker shadow optical illusion video

This is magic: a video demonstration of Edward Adelson's famous checker shadow optical illusion. (via Daily Grail)

• [tmccartney66]

  I’ve seen this illusion before, but not demonstrated on video like this.  That said, there’s something hinky going on with the video.  It looks like the color of the square she’s moving back and forth has been manipulated in some way with Aftereffects or something.  There’s a little video distortion on it as it magically changes colors.

  • [mat catastrophe]

    Don’t know about all that (color stuff) but there’s a painfully obvious jumpcut in there.  (@47 sec)

• [Moriarty]

  Stupid brain.

• [Rob Beschizza]

  I took grabs from two frames on the most obvious and impressive “tile drag” — one at the beginning when the tile appears light and one when it appears dark — and there’s no apparent color change in the tile at all.

• [Blaze Curry]

  Also: nice living room. I want one just like it, but with more rugs.

• [Tino]

  Wait, this is not how it’s supposed to work, is  it? 
  In the drawing the shadowed light square is the same hue as the illuminated dark square. You can check by cutting it out in gimp and move it around.
  Here the the check is moving the actual square from the shadow to the light! This should CHANGE the percepted  hue of it and in no way have the result we see. Is there some image manipulation going on?

  • [PolishQ]

    I think in this case the lights surrounding the checkerboard are mostly misdirection and the majority of the “shadow” is actually painted on.

• [Brent Kirkham]

  It’s to do with the shadow/s.  Look carefully, even with the shadow on the board, all the squares appear to be uniform dark and light grey.  If the shadow were real, the lighter square within the shadow would appear darker (tonally) than those not in shadow.

  As PolishQ says, the tiles have been manipulated in such a way that the lighter tiles in shadow are the same as darker tiles at the edge.

  Sweet illusion though.

• [nosehat]

  Yes, the “shadow” is entirely artificial here, painted on to the squares. 

  You can tell because her arm doesn’t become darkened at all when she reaches into the “shadow”.

  • [emg72]

    I agree about the arm — but if the shadow is painted on, then explain what happens between :45 and :47 — I think you’ll see that the shadow passes over the tile as she pulls it out from behind the cylinder. If it were painted on, that couldn’t happen. 

    I’m also not sure why everyone suspects manipulation or CGI — it’s a real-world example of a commonly understood illusion. Let’s let a thing be a thing, no? Or am I just way too trusting of, you know, the world?

    Also: clearly, she is a witch. 

    • [Guest]

      The “shadow” is mostly artificial (“painted” onto the checkerboard), but the two lights aren’t entirely misdirection.

      I’d guess that having actual lights creates real shadowing–you can see the shadows cast by the witch herself (does that prove she’s *not* a witch?)–in addition to the painted shadow effect, and that makes the illusion more convincing.

• [Bodhipaksa]

  I suspect that the entire structure is CGI. You can see the “shadow” pass over the “tile” as the woman moves it. Taken as a whole, this doesn’t make sense if this is a real object.

• [Laurel McMechan]

  two words: Josef Albers

• [ZikZak]

  To everyone claiming this is shopped – how can you tell?  And anyway, how many shops have you seen in your time?

  • [Antinous / Moderator]

    Is it a local shop?

    • [labrys]

      for local people?

• [Grim Beefer]

  So I just tried to replicate this image myself in 3dsmax on my lunch break, and I’m quite pleased with the results. The two values in the squares pointed two by the two arrows are indeed nearly identical, as any check in an image editing program would tell you. There’s no trick to it, no painted on shadows or anything of the sort. The optical illusion is created by a careful balance of the gray tones of the checker and the output levels of the direct and ambient light. I can post the 3d file, if anyone’s interested in playing around with it themselves, as I don’t have all day to get the results perfect. I have no doubt that this would work in a real life model, as the video demonstrates.

  • [jere7my]

    Beefer: Note the shadow around the edge of the platform in your image, and the lack thereof in the video. The side-light is a red herring; it has an effect, but a small one. If it were as strong as it appears to be, the platform would cast a shadow too. The “shadow” the cylinder casts is indeed painted onto the grid; the scene is illuminated by the overhead light.

    No manipulation needed.

• [Neo]

  Of course the shadow is painted! Actually, the shaded area has squares that were printed with a gradient so that they are partially darker until they match the dark gray in the “lit” squares. If the shadow were NOT painted, the illusion could have begun by showing the grid BEFORE the cylinder is placed in the picture, showing a grid that is even. They cannot, because the grid has been manipulated to pull off the illusion. Check and mate.

• [Neo]

  Just checked the YouTube original site and the makers actually explain the shadows are printed. End of discussion ; ) 

• [LightningRose]

  Anyone who thinks the images are manipulated (ie: shopped) need only cut a couple of holes in a card or piece paper that masks off all but the two relevant squares.

• [Grim Beefer]

  The youtube explanation could do a better job of explaining what they are doing.

  They’re obviously doing some form of manipulation to keep the color of the tile consistent between being in and out of shadow, but this is only for demonstration purposes. In other words, the only reason they manipulated the object was to be able to move the tile back and forth. You could create a “real” version of this optical illusion that needs no such trickery using a uniformly valued checkerboard.

• [catbusrider]

  I loved this illusion when I first saw the static image version-I even had to take it into photoshop and check the values, and even copy pasted the squares before I actually believed it.

  For some reason this video just isn’t as interesting or convincing as that simple static image. 

• [Atrum]

  Witch!  Witch!  Where are my witch-weighing scales and duck?

• [Coal Miki-Restall]

  The light that appears to be casting the shadow is angled upwards. It’s there to give a sense of depth to the upright tube making the shadow seem more natural.