As a mathematician I'm hypersensitive to people overreaching with theory. It's often the King's clothes. There are indisputable reasons why a musical fifth sounds good, but vision is more organic. Evolution has left us with many layers of color perception, like an out-of-control code base. Sure, anything can be fit to a model if we ignore the parts that don't fit; that's how theory intoxicates.
My father devised the "Bayer filter" for digital photography, and favored green for practical reasons that have stood the test of time. There is some theory that applies here, it's just not the same as harmonics and sound.
I'd love to code up a machine learning project that showed the user many color combinations, responding to feedback. Willem de Kooning for example painted with an extraordinarily original palette. Could we tailor our own color palettes, with an AI assist?
All theory is overreaching. It's called the Problem of Induction. All you can do is hypothesize, test, and revise. Mathematicians will never be happy about it.
It will always be worse the more complicated something is. One particle is hard. A hydrogen atom is harder. Oxygen is just this side of impossible. Molecules are impossible and you just start throwing away vast numbers of terms. By the time you get to an eyeball, much less a brain, you are basically just guessing and hoping you can cover some of the cases some of the time.
> I'd love to code up a machine learning project that showed the user many color combinations.
I teach painting in an art school. The huge problem with almost all pallet choosing apps (e.g. Adobe's https://color.adobe.com/) is that they produce swatches: a small collection of discreet color values (e.g. red, green and yellow). These would present as peaks in a hue histogram. These swatches would be great for choosing a color scheme for your kitchen, but are not suitable for a painter.
I have looked at more painting hue histograms than any man alive, and I can tell you that almost all painters employ ranges of colors. The hue histograms present as 'pie slices' on the histogram, not peaks. Hence: a wide pie slice of greens (plural) and s narrow pie slice of reds. Show me a color picker like that and I would likely share it with my students. As it stands, I warn against them.
Most modern dedicated painting apps (Krita as an example) support continuous gamut masking. There are also several third-party pickers for non-dedicated apps (ex. Photoshop) that support it.
> I'd love to code up a machine learning project that showed the user many color combinations
That would be fine, so long as you don't use an RGB display to display those colors.
The RGB color space is severely limited when compared to the colors a person can see. This becomes obvious as one plots color spaces on a CIE 1931 diagram.
Here's a simple tool to see some of this (recommendation: move "system gamut contrast" slider to 50%). Select a working color space from the pulldown to see how much of the visible range it encloses (or discards).
A typical REC709 color space throws away over half the colors a person can see. In other words, as I said, using a monitor to run color experiments can be a serious mistake. Color Science can be strangely complex.
There is a company trying to push the envelope in display technology and encourage industry to move primaries beyond RGB. The results, based on prototypes, can be impressive:
If not under an NDA, what are they using to represent the super-bright reds and greens? Lasers? And, more interestingly, the super-greens and super-blues that are very dark and highly-intense?
(I worked for a printing house and I know that some colors are outside RGB and outside CMYK, so they are printed as spot colors. There is a ton of Pantone colors you cannot mix, and must add from special one-color cans, like the Reflex Blue. Some customers want them badly.)
RGB lasers are already a thing for high-end projection. They have one picture of projectors on their website[1], I'll guess they are modified off-the-shelf ones with custom lasers. You can just barely make out the brand Christie on them, and that is a manufacturer who have laser projectors[2].
Considering their main focus is in the new primaries, it wouldn't surprise me if they just used the standard lasers for red and green, maybe even blue; sourcing suitable high-power lasers in some specific wavelengths might not be that easy.
Besides projectors, the other thing they seem to have done is a "dvled" (direct view) display, stuff thats maybe most familiar as those outdoor signage displays. This press release[3] has one blurry picture of their four primary led display
It's funny that they call their tech "Full Color Range™" while it still catches smaller part of gamut that some industry standard colorspaces (ACES 2065-1) do with three primaries :)
Of course improving capture or display tech is good still, I'm just less convinced of the idea of trying to upend the whole industry
> smaller part of gamut that some industry standard colorspaces (ACES 2065-1) do with three primaries
The ACES 2065-1 primaries are not realizable, they are imaginary. In other words, one cannot make a display capable of displaying these primaries. They are mathematical tools enclosing the entire CIE 1931 space.
Yes, but of course it is debatable how relevant realizability is in practice; we have long history of using displays that have significantly smaller gamuts than whatever nominal colorspace being used.
Was rec2020 realizable in practice when it was created? I'd bet no.
Sorry, I think you might be mixing up a couple of ideas. In this context, not-realizable means “not real”. Rec 2020 primaries have always been realizable. ACES AP0 primaries only exist mathematically for the purpose of enclosing the entire “wedge” of visible colors.
Exactly. It's the best proxy for grayscale detail.
One could have balanced RGB in a hexagonal grid. That's harder to build and slower to process.
He would have considered a hex grid, but our rectilinear cognitive bias was pretty entrenched. Before this work, he programmed an entire image processing system in the "ed" PDP editor, manipulating character arrays. That experience influenced both the Bayer filter and Bayer dithering. And it taught me to look for simple ways to do things.
Theres also the fact that humans have wide variety in vision
Even ignoring color blindness, there's the fact that some people see significantly more range of colours than others (I am one).
Women tend to be more sensitive to slight variations than men (surmised to be evolution ensuring they can better gather from plants , knowing ripeness etc). I see even more variety than that norm.
Human vision is funny, because the majority actually don't see much, with some seeing even less, and some seeing so much they can perceive parts of the UV and Infrared spectrums (not pleasant since our world is absolutely not designed for such people).
It seems unlikely that AI could actually do well accommodating all the variations and would almost certainly be focused on averages while ignoring the color blind and "advantaged".
> There are indisputable reasons why a musical fifth sounds good, but vision is more organic.
Once could argue that our eyes can't even "hear" a fifth. The amount of data our ears extract out of basically two points is amazing. Our eyes in the meantime take an entire spectrum and come up with "blue-ish".
I think you're pointing at the fact that there are only three kinds of cones, whereas the ear has … well honesty I'm not sure how many different kinds of hair cell there are. apparently [16,000](https://www.nih.gov/news-events/nih-research-matters/hearing...)!
the eyes are much better at directionality though; probably some kind of trade-off going on there.
Well it's position vs frequency, so a tradeoff is forced by the Heisenberg uncertainty principle. I don't know if either system is actually reaching this limit though.
I thought about that, but you could just cram more cells to get higher resolution. Reflecting a bit more, hair cells vary by length, whereas each new cone requires a unique protein, so I'd expect hearing to respond to more easily to evolutionary pressures.
> hypersensitive to people overreaching with theory
What’s left of the scientific method if people don’t? I think reproducibility and inertia („science progresses one death at a time“) are problems, but I hadn’t thought that induction could be one.
Maybe off topic, but I'd love to know more about those "practical" reasons for having two green bits in the Bayer system. I guess to your point, they needed four readings and might as well double up on the band humans have more cones to detect.
But also I wonder how those kinds of decisions end up coloring so to speak AI. Training a model on image compression designed to fool the human eye is basically forcing it to reverse engineer the human eye.
I think it is important to
point out that this essay only claims that
Hering’s theory and general concepts of color opponency and cone opponency is wrong and not the idea of opponent colors in general.
"Proof of color opponency was established decades before Hering, with the discovery of complementary-color pairs and color afterimages [29.,30.]; color opponency is implemented by retinal cone-opponent neurons [22.]. Trichromacy and complementarity organize colorimetric space [31.], and Hering’s theory can be discarded without threatening these well-established principles."
> Objective tests of Opponent-Colors Theory became possible when the theory was formulated as a hypothesis about how cone signals are transmitted to perception. The formal exposition spells out the mathematical transformation of cone responses to opponent-color pairs; baked into the math is the linearity of Opponent-Colors Theory implied by the description of color appearances as simple (mathematical) combinations of red-versus-green, yellow-versus-blue, and black-versus-white
Why does the fact that colour appearance can be described in terms of the unique hues imply that the representation must be linear?
This is a good question and I'm also not quite sure what to make of it. As no one has answered this for a while I'll give it a shot, even at the risk of being wrong.
I think that what they are saying is that linearity is part of the status quo of Opponent-Colors Theory and they reject the whole thing including the linearity. So in essence they agree with you.
I have a simple theory of colors. Look at how the color cells are distributed in the retina. Same pattern, but at slightly different scales (spatial frequencies). I speculate color is coded the same as spatial information. Or rather, the brain assumes that spatial frequencies coincident with the cell spacing are most probably because the object is that color, not because the object is weirdly striped as black and white in exactly the same pattern as the retina. But as far as data stream goes, the color and spatial information are the same thing.
This theory comes entirely from observations on (and of people on) LSD. In particular, fine grain spatial information often drifts out of phase and manifests as a rainbow effect. Looking at an image of TV static its possible to see it as the black/white parts swirling or as color glitching outlines depending on how you choose to focus your eyes.
“Color (American English) or colour (Commonwealth English) is the visual perception based on the electromagnetic spectrum.”
⇒ colors aren’t coded; they are created in the brain. Outside it, all there is are light spectra.
> Or rather, the brain assumes that spatial frequencies coincident with the cell spacing are most probably because the object is that color, not because the object is weirdly striped as black and white in exactly the same pattern as the retina.
So, how do you explain that subjects do not change in colour when you move closer to them or further away, or when you divert your gaze (cone cell spacing is far from uniform across the retina)?
> So, how do you explain that subjects do not change in colour when you move closer to them or further away, or when you divert your gaze (cone cell spacing is far from uniform across the retina)?
To be clear, this is not a rejection of the theory, because your brain does a lot of magic to make color (or even object) perception "stick" as things move in and out of peripheral vision. See for example:
Yes, but not to the extent this theory would predict.
If “the brain assumes that spatial frequencies coincident with the cell spacing are most probably because the object is that color, not because the object is weirdly striped as black and white in exactly the same pattern as the retina”, an object perceived as red from 4 meters away would be perceived as purple from 2 meters away (or vice versa), and as yellow from 3 meters distance (the visible light spectrum roughly is from 400nm to 800nm)
No it wouldn't. An actually red object would have the exact same red-cone spatial pattern no matter how far away it is, because the pattern is coming from inside the retina not from the object itself. An actually weirdly striped black and white object with the exact same pattern as the retina would behave like you said.
So you’re saying the data stream to the brain is essentially an analog NTSC TV signal?
Because that’s how color was encoded: as high frequencies in the spatial scan. This was conveniently backwards compatible with analog black & white TV, at the expense of color accuracy (which is why the standard was sometimes called “Never Twice Same Color”.)
Like NTSC/PAL but in a (possibly redundant) basis that is highly optimized by evolution for 3d objects projected down to the 2d retina. Not a rectilinear scanline per se but conceptually similar.
Yes, Margaret Livingstone discusses this in a chapter of the book 'Vision and Art: The Biology of Seeing'. Animals evolved to have black & white vision first and then as color vision evolved the color information was patched onto the black & white signal.
Hasn't this been established for long by references to antique color names (both Greek and Roman), as well as the representation of spectra in African languages? While the article makes reference to Greek semantics and anthropological findings, it still presents this as something new, where it should have emphasized the underlying theory.
This essay should not have discussed anything about the names used for colors in various languages, because the number of color names that exist in a language has a very loose relationship with the psychological perception of colors, so it cannot be used to prove anything about the validity or invalidity of Hering's theory.
In all languages, the circle of hues is partitioned in a number of regions, which have distinct names. The number of these regions varies from language to language, usually between two and eight. The value of this number has much less to do with the distinctiveness of the colors, than with the necessity of describing the colors of the objects that had a high probability of being subjects of conversation for the speakers of that language. Many languages have a single word for the colors green, blue-green and blue, just because there were no common objects for which it was necessary to specify whether they were green or blue.
The too much repeated myth that the Ancient Greeks did not have a word for blue was created by people who might have known some linguistics, but who were ignorant about chemistry and mineralogy.
There are plenty of dictionaries and commentaries about Ancient Greek literature which are very wrong about many words, because their authors did not understand what they have read, for lack of knowledge about the natural sciences. For instance, when Homer speaks about "miltos" used to paint some ships, you must understand that this word designates what is now called hematite or ferric oxide, which was processed as a red pigment. Or else when Plato speaks about the grains of diamond found together with alluvial gold, that has nothing to do with what is now named diamond (which became known to the Greeks only later, after the expeditions of Alexander the Great), but "diamond" was the name for nuggets of native osmium-iridium-ruthenium alloy. Understanding this meaning makes clear why Hesiod said several times that diamond is gray and why he considered it as a material from which a blade could be forged by someone with superhuman strength.
There are many references to blue in Ancient Greek literature and the normal word that was used has been borrowed in English as "cyan". "Cyan" has never meant blue-green in Greek, but only plain blue. "Cyan" initially meant the color of the painting pigment that is now called "ultramarine blue", but later it was also used for the cheaper blue pigments Egyptian blue and azurite. An alternative way to refer to blue in Greek was as the "color of the air", which meant the color of the sky.
In Latin, blue was normally called as the color of the sky, while the word used for green, "viridis" meant either green or blue-green. When these two colors had to be distinguished, the former was described as green like leaves or like grass or like emeralds, while the latter was described as green like the littoral sea or like turquoise or like beryls.
Similarly, the word used for red in Latin (and also in Greek) meant either red or purple, and when the two colors had to be distinguished, the former was described as red like the dye extracted from beetles (crimson), while the latter was described as red like the dye extracted from marine snails.
Auto industry is all based in the opponent-colors theory, imagine convincing thousands of people that the problems and fines involving color quality control were wrong all along.
From what I understand are they not rejecting all of it, just the parts that hypothesized how things work in brain. These parts are hard to verify and I believe the essay claims that these parts have been falsified by now and it proposes an alternative hypothesis.
My father devised the "Bayer filter" for digital photography, and favored green for practical reasons that have stood the test of time. There is some theory that applies here, it's just not the same as harmonics and sound.
I'd love to code up a machine learning project that showed the user many color combinations, responding to feedback. Willem de Kooning for example painted with an extraordinarily original palette. Could we tailor our own color palettes, with an AI assist?