Matthew Cob sent a tw**t by Perfectly Timed:
What’s going on here? With the help of Greg, our Color Man, we can explain it thusly: The white line at the top of the bottom half implies that the light on this hinged object is coming from above, and thus the bottom half of the object is the same color as the illuminated line, and lighter as well than the top half of the object. But it isn’t: it is in fact the same color as the top of the object. We compensate for the assumption that the bottom is lighter by visually interpreting it as lighter, so the object looks as if it were in two shades.
This is one way that evolution, I suspect, has conditioned for vision, making our brains interpret the assumption of shadowing in a way that compensates for it. A famous example of this compensation, which I’ve written about before, is the “checker shadow” illusion in which a cylinder rests on a checkerboard and casts a shadow:
Squares “A” and “B” are actually the same color and shade (if you don’t believe that, go to my original post to see for yourself, as well as the explanation for this remarkable illusion).
I literally shouted an expletive when I put my finger across the page!
I remember getting into a big row with a co-worker in 2002 (approximately) about the checkerboard illusion. They just wouldn’t believe the two squares were the same color, even after I taped a paper across the computer screen and stuck two holes in it.
What a spectacular illusion. The more recent one…eh, not so much.
I had to block the blue and the brown before the colors looked the same.
If you un-focus from it and just gaze at it you can kind of see they’re the same, or at least I can. It doesn’t work for the second one though. That’s just too damn good.
Again, to be fair, in any real-world scene that would form a similar image on your retinas, the intrinsic colors would be just as different as you perceive them. The amazing thing isn’t that we’re not so good at determining absolute illuminance levels, but that we’re so astoundingly good at determining the actual abstracted ideal surface properties of objects.
Cheers,
b&
This immediately sounded wrong to me. And when I thought about it a bit, it still sounded wrong. And the more I thought about it, the more convinced I was that you were wrong. Eventually I googled “real life checker board illusion” and this youtube video came up…
http://m.youtube.com/watch?v=z9Sen1HTu5o&desktop_uri=%2Fwatch%3Fv%3Dz9Sen1HTu5o
Enjoy!
Yes, I’ve posted that video before — even in the last illusion thread we had here.
You’ll note that the light that appears to be casting the shadow is, in fact, pointed upwards and not actually casting a shadow on the board. Instead, the board is evenly illuminated from the softbox. Just as in the static version, the shadow is painted on, and the whole thing is a trick of perspective and camera angle and the like, the same as those sidewalk paintings that look like you’re about to step off the curb into the pits of Hell — but only from the proper camera angle. Move to a different position and everything’s quite distorted.
In that video, if you were to turn off the theatrical lighting and remove the cylinder, the illusion would disappear entirely.
Cheers,
b&
Okay, perhaps I misunderstood what you meant.
The two squares are exactly the same colour as the host demonstrates by picking up the first square and placing it over the the second square. But in its orginal position the two squares appear to be entirely different colours.
In other words, the analysis of the 3D version (which I presumed you meant by “real life”) is identical to the 2D version.
So I guess I’m wondering what the point ofyour comment was.
My point is that, in the movie, the “shadow” is painted on in exactly the same way as in the original illustration. All they’ve done is printed out the original illustration and photographed it in even illumination with a couple extra props.
The video and the illustration are exactly the same. And you’re looking at both on the exact same computer screen. The only difference is that one is a moving picture and the other isn’t.
And both only work from a single perspective. Change the perspective of the illustration (by blocking out the cylinder) or on the stage where they shot the video (by walking around), and the illusion vanishes as quickly as one of these sidewalk paintings viewed from the worng angle:
http://www.boredpanda.com/5-most-talented-3d-sidewalk-chalk-artists/
Cheers,
b&
This is what you said:
” any real-world scene that would form a similar image on your retinas, the intrinsic colors would be just as different as you perceive them”
This is incorrect.
And you seem now to be agreeing that it is incorrect.
For a start, there is no such thing as intrinsic colour. Colour depends on many variables including ambient light and angle of reflection.
You seem now to be agreeing with this.
And all this applies to the real world.
Hence my confusion about your earlier post quoted above.
<sigh />
And the computer display on which you’re viewing all this is in the real world, too, so there’s no need to go nearly so far if all you’re after is a “Gotcha!” moment.
I’ve made it quite clear in multiple posts that what I here used the word “intrinsic” for refers to standardized viewing conditions. And I’ve made multiple posts using the analogy of comparing the readings from a contact spectrophotometer with those from a spectroradiometer.
I hope you’ll forgive me if I don’t recapitulate all that in this response. If you read up all my previous responses on the subject, in this thread and the others, you’ll hopefully get at least the gist of my point.
Cheers,
b&
It seems, then, that you agree that your original statement regarding the difference in the “real world” was incorrect.
That’s all I was confused about, so thanks for clearing that up.
Perhaps I just didn’t understand your original post thought it seemed clear enough that you were implyimg that the real world is different from the visual illusions we see illustrated on the net.
Even cooler is that we all have to learn to see. The image that first strikes our sensors as infants needs lots of processing before objects can begin to be discerned. There’s a lot of fumbling around before something as simple as a cube can be read efficiently, before we learn to read one face as up and another as forward and learn to see those areas of shading as transitions from one surface to another. It’s likely that there was a time when we read both surfaces in the illusion as the same value.
Even cooler than that is that before even that our brains must learn to not see all the glop in front of our sensors. Stuff like nerve fibers and blood vessels actually cover the rods and cones which react to light and they must be discounted before the relevant information of edges and boundaries can start to be distinguished.
I loved Margaret Livingstone’s book, “Vision and Art: the Biology of Seeing” which has a number of these illusions, some which point to real anomalies in our visual system. She also gives a good overview of the various levels of visual processing.
I hate this because it makes my brain hurt.
I love these. I use these kind of examples to show people that our brains are always making the world look better and simpler than it really is. As in the movie “The Matrix” I call the picture and the rest of our representation of the world to ourselves, the “construct.” If you ever have to design robot visual systems, you will quickly find out how difficult making such a construct is, and how little we notice all the things it is constantly doing for us.
Or you could just take a class with my Russian painting teacher, who had a sense of colors and values the way some musicians have perfect pitch. Just about every class would contain a demonstration of these sorts of illusions, except he could show them to you in every still life set-up, showing you, for example, when the the light color in shadow was actually darker that the dark color in light – which wasn’t what you’d first assumed at all.
I would say rather that our brains try, usually with success, to make the world look like it really is. Optical illusions are the cases, not generally encountered in reality, in which the mechanisms we use to correct the image cause errors. In the real world, objects in sunlight are darker than they appear, and objects in shadow are lighter than they appear. It’s really useful to be able to see the same color (meaning the reflectance characteristics of a surface) despite differences in illumination.
This is wrong on so many levels it’s hard to know where to begin.
Our brains do not try to make the world look as it really is. It simply tries to perceive the world in such a way as to improve our chances of survival. Contrasting objects from surrounding objects is useful for survival (so that we see the tiger amongst the shrubs). Both of the above illusions result from this tendency of our brains to contrast objects from their surrounds.
Also, optical illusions abound in nature, we’re just mostly unaware of them. Just like no one is aware that the above illustrations are illusions unless it’s pointed out to them.
And the mechanisms do not attempt to correct the image. On the contrary, they manipulate the image to the host’s advantage (to more clearly see that tiger)
And whatever do you mean by “objects in sunlight are darker than they appear”? The way objects appear is the result of both the wavelengths of light they reflect and the intensity of the illumination. All objects appear black in zero illumination (just turn off the lights). And all objects appear white in intense illumination. In mid morning sunlight, different objects will appear differently coloured depending on many variables such as the wavelength of light they reflect, the surrounding objects, their angle of inclination to the observer etc etc.
A more useful thing like this would be one demonstrating that your particular savior is an illusion.
Done!
http://www.eyetricks.com/jesus.htm
I’m confused by the “white line” versus “the boxes are the same color.” Is this just semantic or am I missing something?
Hard to tell, because you don’t say what confuses you. The “white line” is the top edge of the bottom box. “The same color” refers to the center areas of the boxes, not their edges.
…except that the ‘color’ of that bottom box is implied by that white line; a more accurate question would be “are the gray areas the same color in both boxes?”
Or, even better: what are the absolute luminance values of the two boxes at the position of the observer, and what are the reflectance characteristics of the two boxes under the same idealized viewing conditions?
The absolute luminance values at the position of the observer are identical, but, if this was a real-world scene, the reflectance characteristics under idealized conditions would be exactly how you perceive them to be.
b&
Didn’t I just say that? 😉
That top one is a new one for me–lots of fun!
Reblogged this on aperi mentis.
Reblogged this on aperi mentis.