Lets talk about the dress

[Indigo]

Okay, so you guys have probably seen this all over the internet by now.  I thought this would be easily explainable through color contrast theory.  But try as I might, I simply cannot get myself to see the black and blue.  I've tried dumping it into Photoshop to change the background color, change the exposure, inverse the colors, absolutely everything I can think of - but I still can only see the dress and white and gold.

If you're not familiar with what I'm talking about, it's all about this dress:


Some people absolutely see this dress and black and blue, and others see it as absolutely white and gold.  I'm in the latter camp.

If you don't believe me that this is even a debate, it's even on CNN
http://www.cnn.com/2015/02/26/us/blue-black-white-gold-dress/

So three questions:

• What colors do you see
• Why are we perceiving them differently
• Is it possible to reverse how we percieve it

EDIT one explanation:
http://www.wired.com/2015/02/science-one-agrees-color-dress/

[ptoing]

Some facts: The actual dress is black and blue, the picture was taken with a really shitty camera of phone.

As far as the colours that I see in the picture: Brown and light blue. If you do not know that the colours in the picture are totally of then the light blue could be seen as reflected light or something.

The actual RGB values when checked are brown and desaturated blues.

I personally do not see how anyone could see the brown parts as black in that picture.

As far as photoshop goes. If you do levels on it and use the black value colour picker and select the darker browning colours you will get what the dress actually looks like IRL, black and blue. That does not change the fact that there is nothing that is black in the image.

[yaomon17]

https://vine.co/v/O21nlFP9KKl
Tilt your monitor.

[Indigo]

As far as the colours that I see in the picture: Brown and light blue.

I'm not asking what literal colors you see - That's easily brown and light blue.  I'm asking what color identity you perceptually see the dress as.  And a follow up - what adjustments can be done to the image to flip your perception of that color identity - ideally without altering the colors of the dress itself, but just the context (because it seems to straddle the very thin line for what people perceive).

[ptoing]

yaomon: That does not work on good monitors.

Indigo: Dull brownish gold and white/maybe light blue.

What adjustments can be done, I don't know. And I don't know if I care. I see the image as it is and make no assumptions about what colours the dress might actually be, because I do not see how that is important. It's a shitty photo of a dress. It's nothing I will lose sleep over.

[Indigo]

Indigo: Dull brownish gold and white/maybe light blue.

What adjustments can be done, I don't know. And I don't know if I care. I see the image as it is and make no assumptions about what colours the dress might actually be, because I do not see how that is important. It's a shitty photo of a dress. It's nothing I will lose sleep over.

You're totally missing the art implications.  Understanding how people perceive color - especially in situations like this where people get such drastically different results - is EXTREMELY valuable to understanding art.  You're simply being mr contrarian again and being too literal.  You, as an artist, totally understand the value of color perception and color relationships.

Edit:
One interesting thing about the colors, which may or may not be related to how people perceive the color identity, is the fact that the colors of the stripes are nearly PERFECT inverses of eachother.  Invert the image, and the stripes are nearly the same.

Here's an image to simplify the discussion:

[yaomon17]

wut you say about me monitor m8?
Anyway, this was posted by ChrisConlon on reddit:


Dr. Stephen McLeod thinks it is similar to colorblindness tests of red-green deficiency, just a lot more sensitive. Like this type of thing:


And of course, there are other varibles like maybe monitor quality and room lighting.

[ptoing]

Indigo: I am not trying to be contrarian. The thing is, that if I do not percieve something in a certain way, I will not try and guess what people will percieve, in how I make my art. Of course there are things like relative contrast and all that which I do take into account and like to play with. But that is neither here nor there in this case.

I can not force myself to see an RGB value that is not there (even though the actual dress is that colour).
I give you that it is interesting that some people seem to be able to extrapolate the actual colour of the dress from this photo, but it would be nice to get some actual scientific stuff on this. Not people guessing.

What I am saying is, that as long as I can not somehow flip a switch in my brain to change my colour perception. And if I can not do that I wont try to guess what other people might see when they look at my art to actually do stuff like this intentionally.

Also, please do not always just assume that I am being contrarian for the sake of it just because I might not happen to agree with you. It is really intellectually lazy.

yaomon:

That comparison is a good example of relative colour perception and how things are influenced by light and shadow.
The actual dress is a lot more blue and black though than the "black dress" there, which I would still say is very dark brown and mid blue.

Some people have reported that when they showed people the dress on the same monitor, they got different results. So while that might play a role, I think that how people are "trained" to see colours play a bigger role.

[Probo]

yeah it looks like a white and gold dress in a cold, dim light source to me. people who see black and blue are colour blind right?? or am i???

[surt]

I don't get it. I think I'll just put this down as another reason to despise humanity/internet.