Luminous Landscape Forum
Raw & Post Processing, Printing => Colour Management => Topic started by: LenR on February 28, 2015, 11:42:32 am
-
I saw white and gold (in what looked like cool shadows).
-
This is one of the dumbest discussions of color in recent memory. I saw a number of sites showing the dress differently AND the bkgnd elements too, indicating not only is this a color management nightmare of image posting of untagged data, the RGB data is all over the map too. Optical Illusions as the Nightly News suggested? A group of awful, different, un-color managed images?
If any good comes of this silliness, maybe the masses who never considered that what they see from images on the internet can be horribly mangled will open their eyes to how unreliable the web is for posting awful images that are awfully produced can appear.
-
C'mon now Andrew, tell us what color you saw the dress to be.
Remember, there.s no wrong answer:)
-
This isn't an issue of color (mis)management at all. It doesn't matter if it is tagged, mangled, etc. What matters is that two (or more) people, standing next to each other, looking at the same image, see it differently. That's what is fascinating about it. Being grumpy and dismissive isn't.
-
C'mon now Andrew, tell us what color you saw the dress to be.
Of course you're joking right?
First off, as I said, I've seen several different posts that all appear different due to what has to be processing and mishandling of the data and probably lots of other factors. 2nd, the color from each individual image you pick is easy to define; open in Photoshop, guess at a profile and examine the Lab values. Those are the color values, not the colors I necessary see per se. Color isn’t a wavelength or property of light. Color, is a perceptual property, something that occurs deep inside our brains.
Are the two color patches (A and B) seen here the same color? They appear different:
http://www.takegreatpictures.com/photo-tips/software-tips-and-techniques/color-management-and-your-display-by-andrew-rodney
This digression about the color of TS's dress is silly and illustrates that the web masses talking about it may have now (?) discovered the need for managing digital color!
-
What matters is that two (or more) people, standing next to each other, looking at the same image, see it differently.
You mean just like this:
(http://digitaldog.net/files/WrapColor.jpg)
BFD!
-
I kept seeing a black and blue dress.
But the way it's been shown in the media on TV has some versions lightened which still didn't change the color of blue and some TV coverage would show the same dress but in white and gold, but I didn't know if they were doing it as a demonstration of what others were claiming to see.
But when you find out the increase in sales of that dress from that particular dress designer/manufacturer, I think we all can figure out it was just a clever publicity stunt.
This has nothing to do with color management.
-
You mean just like this:
No, I do not mean "just like this." I meant looking at the same, single image and seeing it as vastly different, not just slightly changed shade.
P.S. Too early in the thread for your profanities.
-
Chartreuse and mauve
-
I think the story is extremely important and might have a very useful impact on the masses. It might popularize the fact that how things look to us is often a matter of perspective and perception. If our very own eyes play tricks on us and lead us to different and erroneous conclusions, it might open our eyes (pardon the pun) to how not just visual, but also cognitive biases shape our opinions. It reinforces the already known concept of unreliability of witness testimony. It might make people pay more attention to how things look like from somebody else's perspective. It might shake the confidence of those who think they are absolutely right in their assumptions, beliefs and prejudices.
-
I meant looking at the same, single image and seeing it as vastly different, not just slightly changed shade.
It is the same image and the same numbers. Just like the optical illusion you missed.
-
No, I do not mean "just like this." I meant looking at the same, single image and seeing it as vastly different, not just slightly changed shade.
If we were all looking at the same physical manifestation of the image, then yes it would be about different people perceiving the colours differently.
But that's not what's happening. The image is being presented as RGB data of unknown colour space, and being rendered on a zillion monitors, most of them displaying the wrong colours.
It's much, much more likely to be about colour management than about colour perception.
My guess, what we're seeing is:
- The white looks blue on monitors with high colour temperature white.
- The gold looks black on monitors with excessive contrast, or being viewed in high ambient illumination.
-
If we were all looking at the same physical manifestation of the image, then yes it would be about different people perceiving the colours differently.
But that's not what's happening.
Exactly! Further, anyone who wished to watch the silly report last night on NBC news should see what I saw, different representation of the same image that looked vastly different in the background part of the image where NO dress represents the image. It's like two totally different manipulated images. Of course they look different!
It's much, much more likely to be about colour management than about colour perception.
colour mismanagement.
Get real folks, just LOOK at just this screen capture (color managed) of the SAME dress from two different sites, do they match?
(http://digitaldog.net/files/dressed1.jpg)
-
http://www.nbcnews.com/nightly-news/video/the-science-behind-the-black-and-blue--or-white-and-gold--dress-405695555960
Just stop the video at around 42 seconds. Slight difference in exposure of the THREE renderings shown? You bet.
Kind of shocking that an audience usually so sophisticate about color, let alone image manipulation would even discuss such silliness. :'(
-
If we were all looking at the same physical manifestation of the image, then yes it would be about different people perceiving the colours differently.
But that's not what's happening...
That IS exactly what is happening.
What you see on TV is a reporting about the phenomenon, not the phenomenon itself. TV end media are reporting about it AFTER it reached a viral status in social media. By the time you see it on TV, you are already a century behind the real event.
This is how it strarted (well, in my case at least): my teenage daughter comes to me with her friend and they show me a single image of the dress on one phone and ask me what I see. I say black and blue. My daughter agrees, but her friend is in shock as she swears she sees it as white and gold. That is what it is about.
-
.... Slight difference in exposure of the THREE renderings shown?...
Jesus, it is like talking to a deaf person. It isn't about looking at two or three renderings, it is about looking at a single image.
-
Exactly. This is not about colour management.
-
I've seen that same image as white/gold and then blue/black. The white/gold image was at the top of web news story. I scrolled down to read the story. When I scrolled back up the dress had switched to blue/black. I was so shocked I thought web designers must have switched the jpeg. They hadn't.
-
The BBC invokes Wittengstein to sort this out:
http://www.bbc.co.uk/news/blogs-magazine-monitor-31662317
-
http://www.nbcnews.com/nightly-news/video/the-science-behind-the-black-and-blue--or-white-and-gold--dress-405695555960
Just stop the video at around 42 seconds. Slight difference in exposure of the THREE renderings shown? You bet.
Kind of shocking that an audience usually so sophisticate about color, let alone image manipulation would even discuss such silliness. :'(
More scientific analyses are given on Wired (http://www.wired.com/2015/02/science-one-agrees-color-dress/) and the New York Times (http://www.nytimes.com/2015/02/28/business/a-simple-question-about-a-dress-and-the-world-weighs-in.html).
Important considerations are color constancy and discounting of the illuminant and these two color scientists do not regard the exercise as trivial.
Bill
-
Jesus, it is like talking to a deaf person.
And this is like talking to a blind one. Are you trying to tell us that just the two examples below I posted match?
The exposure examples shown are all over the planet in just the video introduction. JUST the one example of supposedly the same image I posted from the web don't look the same that supposedly show the 'same' black and blue dress upload. This is mangled image data looking different because it IS different. The screen capture I posted below isn't different due to an optical illusion, the data is different or treated differently.
-
The fact that images distributed around the web end up looking different is a red herring. (Or is it blue?)
As Slobodan has stated, the phenomenon of the white/gold/blue/black dress is about several people making different interpretations of the SAME image. On the BBC News the roving reporter was showing people the image on his iPad. Some would see a white/gold dress, others blue/black.
-
http://www.wired.com/wp-content/uploads/2015/02/Untitled-12.jpg
From Wired. Ignore the dress (hold something up to block it), view the bkgnd and that red item and surrounding area to the right of the sleeve. They don't appear the same? They are not the same. Got nothing to do with a dress. The exposures are not the same for one.
(http://www.wired.com/wp-content/uploads/2015/02/Untitled-12.jpg)
Take this untagged file from Wired into Photoshop, assign sRGB and place a sampler over that red as I did. I get the following values (left to right):
R255/G203/B155
R252/G130/B119
R255/G75/B87
Optical Illusion? No, difference in RGB data and exposure. Now take the optical illusion I provided in my article below and do the same. Same RGB values.
Some of you are being duped as fools with this nonsense. At least as provided by the web of these various images.
-
When I scrolled back up the dress had switched to blue/black.
Well I see the image on Wired visually change by simply moving my cursor into the web content so what's that all about? It's NOT an optical illusion.
-
Jesus, Andrew, I knew you are dense, but not this dense. It isn't about "moving the cursor" or two or three images seen either simultaneously or in sequence, it is about seeing a single image, on a single device, by different people. How much simpler can I expain that!?
-
Andrew, you've posted Wired's trio of images, but I'm not sure you understand what they are. The central image is the image circulating around the web and causing all the confusion. The images on the left and right are obviously different. They've been photoshopped by the Wired photo editor to represent a white/gold dress and blue/black dress. These two images are not the image causing confusion. Just the central one.
-
Jesus, Andrew, I knew you are dense, but not this dense. It isn't about "moving the cursor" or two or three images seen either simultaneously or in sequence, it is about seeing a single image, on a single device, by different people. How much simpler can I expain that!?
Provide me a means to see the image whereby I view a visual difference without a numeric one will you? I've seen none provided as yet. While the optical illusion I provided does exactly that AND is obviously a real optical illusion. I'm not saying optical illusions do not exist, JUST THE OPPOSITE by providing them over the years and again today. I'm stating that every attempt to show this dress image is done with differing images that look different because THEY ARE DIFFERENT. I'm also not suggesting an optical illusion of said dress is impossible! None has been provided. Every example shows different colors because the colors ARE different.
-
Provide me a means to see the image whereby I view a visual difference without a numeric one will you? I've seen none provided as yet. While the optical illusion I provided does exactly that AND is obviously a real optical illusion. I'm not saying optical illusions do not exist, JUST THE OPPOSITE by providing them over the years and again today. I'm stating that every attempt to show this dress image is done with differing images that look different because THEY ARE DIFFERENT. I'm also not suggesting an optical illusion of said dress is impossible! None has been provided. Every example shows different colors because the colors ARE different.
Er, he's already given you an example. Him, his daughter and her friend all looking at the same image on a phone and seeing it differently. That is the story.
-
You mean just like this:
(http://digitaldog.net/files/WrapColor.jpg)
BFD!
I don't get this. What is it demonstrating? The images look very different to me. Are you going to tell me they're the same?
-
Er, he's already given you an example. Him, his daughter and her friend all looking at the same image on a phone and seeing it differently. That is the story.
So you and he can't provide ME that experience? Telling.
Trust but verify.
Meanwhile, my optical illusion passes the smell test and, every example of this dress image I've seen has differing values. Again telling.
-
So you and he can't provide ME that experience? Telling.
You can easily verify the test with your friends. Try it.
-
I don't get this. What is it demonstrating? The images look very different to me. Are you going to tell me they're the same?
Yes.
-
Yes.
Ok. They look very different to me. The numbers in photoshop are very different. In what way are they the same?
-
One has the wrong embedded profile (initially).
-
Right. That's not what this story is about. It's about people seeing the SAME image differently. Radically differently.
Your approach to this story is a good example of how blinkered 'experts' can be.
-
Right. That's not what this story is about. It's about people seeing the SAME image differently. Radically differently.
That's what we're being told... As yet, many of the image posted looks different and ARE different. I've illustrated this already. Meanwhile, the concept of an optical illusion, one no color savvy user would ignore and something most here should accept and understand has been posted and the colors are identical.
Is the idea that an optical illusion of color something new to the OP or some of you here, that make this silly dress new worthy to you?
-
Here's the BBC reporter out on the street, with the image on his iPad (this might not be viewable outside the UK):
http://www.bbc.co.uk/news/uk-31655236
5 people say white/gold. 5 people say blue/black.
Cockney chap to reporter: 'Are you turning my spanner?'. Same question to Andrew.
-
I see it blue (very light) and gold (dark). no white or black anywhere (white, black or grey should be completely neutral IMO)
The attached image shows a pure white and black patch (255,255,255 - 0,0,0) to the left and two patches with RGB values sampled from the dress. I then tested varying the exposure from -4EV to +2EV (after that the blue patch clips to white) and still looked blue and gold to me.
-
I see it blue (very light) and gold (dark). no white or black anywhere (white, black or grey should be completely neutral IMO).
Me too, it's the only way I've seen it within the vastly different renderings from differing sites. My wife see it that way too. I asked the three Whippets what they see since dogs have different color vision than humans. They showed total non interest in the image as I suspect most people should have in the first place.
Meanwhile, the somewhat famous Adelson optical illusion has been consistent among the hundreds of workshop precipitants I've shown it to over the years. No one seeing for the first time have even reported NOT saying that A and B appear different prior to showing the 'proof' they are in fact the same color. The numbers have never been different either.
http://web.mit.edu/persci/people/adelson/checkershadow_illusion.html
-
Sorry but I thought it might be a fun conversation.
I didn't expect the Spanish Inquisition!
Me thinks there might be a little lightening up called for here:)
-
So you and he can't provide ME that experience? Telling...
Yes, Andrew, for a change you are right this time. Neither I nor Elliot nor anyone else at this moment can provide you THAT experience for a simple reason: you've come late to the party. You've already seen several versions of the image, some color corrected, some exposure corrected, some single, some in double or triple comparisons. Your perception at this moment is forever polluted by what you've seen and there is no way you can UNsee it now. Seriously. And anyone else who has not seen what I saw (and those in Elliot's BBC example), i.e., the original, single image, that I (and others) saw for the first time in our lives, can not possibly understand what the fuss is about now.
-
... Meanwhile, the somewhat famous Adelson optical illusion has been consistent among the hundreds of workshop precipitants I've shown it to over the years. No one seeing for the first time have even reported NOT saying that A and B appear different prior to showing the 'proof' they are in fact the same color. The numbers have never been different either.
http://web.mit.edu/persci/people/adelson/checkershadow_illusion.html
Andrew, I've been aware of the checker shadow illusion for quite some time. And yes, you are right: "No one seeing for the first time have even reported NOT saying that A and B appear different." That is why this dress "thing" (perhaps calling it "illusion" is not entirely correct) is so fascinating, because different people do see it differently. Another difference... for instance, there is this famous and old black & white illusion, one where some people see a black vase in the middle, and some other people see two white portraits on the sides. However, after a while, or if pointed out, people will eventually recognize the other pattern. Not in the case of the dress: even after realizing that their friends see, say, black/blue, the other side will fight to death that it is white/gold.
-
That is why this dress "thing" (perhaps calling it "illusion" is not entirely correct) is so fascinating, because different people do see it differently.
I don't find it fascinating because there all kinds of optical illusions that have exhibited this 'effect' (if it is even real with the dress, I'm not at all convinced) for years. That some here haven't see this before and find this silly dress anything new under the umbrella of color is a bit fascinating and a bit sad. For teenage girls that follow TS, I can see how enlightening and exciting this all must be.
http://www.funzug.com/index.php/unusual-things/great-optical-illusions-magic.html
http://www.planetperplex.com/en/subcategory/optical-illusions/color-illusions/l/
http://www.washingtonpost.com/blogs/wonkblog/wp/2015/02/27/12-fascinating-optical-illusions-show-how-color-can-trick-the-eye/
-
This is how it strarted (well, in my case at least): my teenage daughter comes to me with her friend and they show me a single image of the dress on one phone and ask me what I see. I say black and blue. My daughter agrees, but her friend is in shock as she swears she sees it as white and gold. That is what it is about.
Similar for me. My son in law ask me last night what color I saw which was brownish black and a muddy blue. When his wife said she saw white and gold I was surprised. The paradox to me was white and gold seem like light colors. He passed the phone around and 5 saw it black and blue, 3 saw white and gold. And none of us could see the other version even though we tried so it wasn't about being an optical illusion. Normally I wouldn't think much of it, but the difference between blue and black vs white and gold seems so extreme. Somehow the visual response or interpretation of what was being seen was involved. But everyone looked at the same device and image.
-
Your perception at this moment is forever polluted by what you've seen and there is no way you can UNsee it now.
Sorry, I'm not buying that. That isn't the case with the Adelson example (and other's I've just posted), A and B always look visually different and I know the trick! As do you. The same is true for how you and I adapt to white. We can't turn that off. You may view one paper under an illuminant then another and tell us they appear the same color, but you can't not see a difference (when it does exist) when you view the two at the same time.
-
I didn't expect the Spanish Inquisition!
That topic was funnier when done by Monty Python and more interesting when not. ;D
-
I don't find it fascinating because there all kinds of optical illusions that have exhibited this 'effect' (if it is even real with the dress, I'm not at all convinced)
I'm not sure what mean by 'not convinced.' Do you think people are lying about what they are seeing? I can't see the image as anything other then washed out black and blue, but clearly some people do. That this isn't like other optical illusions where the illusion demonstrates perceptual conflicts that are consistent from person to person is part, if not most, of the fascination. I've spent quite a bit of time with the color perception literature, and I can't think of another example where people consistently report such dramatically different color perceptions of the same stimulus. That's why this is interesting, at least to me.
-
The dress is blue and black, opinions of the colors in the image may differ, but the dress itself was blue and black.
Brian A
-
I'm not sure what mean by 'not convinced.' Do you think people are lying about what they are seeing?
I have no idea what they are seeing. I've seen so many totally different images of the dress, obviously using different numeric values, due to who knows what manipulations.
-
I have no idea what they are seeing. I've seen so many totally different images of the dress, obviously using different numeric values, due to who knows what manipulations.
Most reports I've seen including the report from Wayne above, and the BBC video posted above, involve people looking at the same image on the same device. That's what all the fuss is about. That there are now a bunch of variations of the image floating around is somewhat beside the point.
-
Mark, as I said, I have no idea what these people were shown or are seeing. I've never said anyone is lying or that differing people might perceive something different. In my very first post about this dress, I said: Optical Illusions as the Nightly News suggested? A group of awful, different, un-color managed images?
I also wrote, and proven I think: I've seen several different posts that all appear different due to what has to be processing and mishandling of the data and probably lots of other factors.
I never said this isn't some kind of optical effect or illusion. I said it's a really stupid discussion among a group outside of teenage girls and maybe others who maybe never knew optical illusions existed and a surprise that this group would find it at all worthy. I also wrote: I'm not saying optical illusions do not exist, JUST THE OPPOSITE by providing them over the years and again today. I'm stating that every attempt to show this dress image is done with differing images that look different because THEY ARE DIFFERENT. That makes evaluating what's going on here more difficult and suspicious.
It's like the one web site that showed the dress with different white balance settings to illustrate the images appear different. DUH! This is news to some on LuLa?
-
Why not explain to us stupid girls what's going on?
-
Why not explain to us stupid girls what's going on?
Some girls are too stupid to read and comprehend, the answer to that question was stated in post #48 and 50. ;D
-
... a surprise that this group would find it at all worthy...
Why it is worthy to some in this group is well explained by Mark:
... That this isn't like other optical illusions where the illusion demonstrates perceptual conflicts that are consistent from person to person is part, if not most, of the fascination. I've spent quite a bit of time with the color perception literature, and I can't think of another example where people consistently report such dramatically different color perceptions of the same stimulus. That's why this is interesting, at least to me.
-
Mark's point is valid Slobodan which is why extraordinary claims require extraordinary evidence.
-
Some girls are too stupid to read and comprehend, the answer to that question was stated in post #48 and 50. ;D
Ok. Let me help you. They're looking at this image in a colour managed browser:
http://www.bbc.co.uk/news/blogs-magazine-monitor-31662317
Some people see a blue/black dress, others see a white/gold dress.
Why?
-
Some people see a blue/black dress, others see a white/gold dress.
Not blue/green? Because I just showed THIS link you provided (http://www.bbc.co.uk/news/blogs-magazine-monitor-31662317) to my wife and she said green, not black. And yes, on a color managed browser! I agree, the upper right side of the 'black' dress has a greenish appearing cast.
Why?
I replied to that question 3 times.
-
No, green is not an option.
-
No, green is not an option.
I see, so what someone else perceives of the site you just posted has to fall into the colors sets you demand?
I'm simply reporting what she said she saw. Just like those reporters. I didn't ask anything other than "What colors do you see" and her answer was green and black.
-
I see, so what someone else perceives of the site you just posted has to fall into the colors sets you demand?
I'm simply reporting what she said she saw. Just like those reporters. I didn't ask here anything other than "What colors do you see" and her answer was green and black.
Ok, she can have green. It's a literalist response which probably comes from being married to a colour geek.
The majority of people will see either a blue/black dress or a white/gold dress.
That's what I'm asking you to explain.
-
I know what you are asking me to explain. I've answered several times. I have no idea what these people are seeing.
-
I know what you are asking me to explain. I've answered several times. I have no idea what these people are seeing.
Ok, fair enough. You can't get your head around it.
-
Ok, fair enough. You can't get your head around it.
Currently, my head's being used to work on a color project that feeds the wife and dogs.
-
Not blue/green?...
Actually, the BBC article did allow the third option: blue - olive green.
But here is the kicker: the image in the BBC article TO ME TODAY isn't the one I saw yesterday. Yesterday, I said blue/black, today I say, looking at the BBC image, white/gold. Now, there is no way for me to know if that is because the image BBC used is indeed a different one, or my mind has already been "polutted" by all the other images in the meantime. The only thing I am certain of is that yesterday, when I saw the image for the first time in my life, I said blue/black, and the other person in the room said white/gold.
-
Actually, the BBC article did allow the third option: blue - olive green.
But here is the kicker: the image in the BBC article TO ME TODAY isn't the one I saw yesterday. Yesterday, I said blue/black, today I say, looking at the BBC image, white/gold. Now, there is no way for me to know if that is because the image BBC used is indeed a different one, or my mind has already been "polutted" by all the other images in the meantime. The only thing I am certain of is that yesterday, when I saw the image for the first time in my life, I said blue/black, and the other person in the room said white/gold.
Interesting. For me it went the other way on the BBC site. First I saw white/gold. Now I'm seeing blue/black. I don't believe they're switching jpegs. If I leave it a few hours before revisiting the image, my initial impression is white/gold, but in seconds it turns to blue/black.
One thing I'm fairly sure of - it has nothing to do with rods and cones (the BBC roving reporter's theory).
-
Whatever you two BBC fan's are smoking, please do share ;D
-
Jesus, it is like talking to a deaf person. It isn't about looking at two or three renderings, it is about looking at a single image.
Some people describe the colours differently, and some people say they see the same picture differently at different times. Yes, of course. Welcome to the world of colour perception. Our eyes adapt to the context around the image, to what we've been looking at immediately before we see the image and even to our preconceptions of the image subject matter.
However, as the Wired article (http://www.wired.com/2015/02/science-one-agrees-color-dress/) points out, it's also about different renderings of the same picture, or viewing the same picture in different lighting contexts.
The fact that a black dress can appear gold at all is likely to be because the image being circulated on the web is poor. As the Wired article shows, in that image the dress is gold (the pixel RGB values are not a neutral grey or black), whereas in real life it's black. Of course, some monitors will incorrectly render that colour, or show it so dark that it is perceived as black.
I find it rather telling that news articles about it refer to philosphers rather than colour scientists.
-
Here’s my explanation.
It’s an over-exposed photo with a tungsten cast. Many people intuitively understand this and describe the dress correctly as blue and black.
The people who see the dress as white and gold have misread the photo. They’ve correctly understood that the photo is not an accurate representation of the dress. They know that something is wrong with the colour and tone of the picture. But instead of seeing an over-exposed photo with a warm cast, they see an underexposed photo with a blue cast, and hence a dress that is white and gold.
What makes this misreading possible? I think it’s the blown-out, flared background. In recent years the flared-out/backlit portrait has become a popular trope in advertising and wedding photography. Normally such pictures are shot outdoors and into the sun. Often the subject will have a blue cast (from the blue sky). The recent ubiquity of such images means that we have learnt to intuitively colour correct the blue cast that many of them contain. That’s what’s happening in this picture. The blue is converted to white, and the ‘black’ (which is already warm) is rendered as gold.
-
The people who see the dress as white and gold have misread the photo.
I don't think so; it's just a poor photo.
In the photo I saw (http://www.bbc.co.uk/news/blogs-magazine-monitor-31662317 (http://www.bbc.co.uk/news/blogs-magazine-monitor-31662317)) the dress was a sort of brown. Call it gold if you like, but it was not black. The "white" of the dress was very definitely blue in that particular picture, by comparison iwth the white background of the web page.
Here's the colour of a sample of the lighter area of the dress in the upper central area of that Beeb link:
(http://www.simongarrett.co.uk/Dress_sRGB.jpg)
The colour (in Photoshop) is R=121, G=107, B=68. (That's assuming the image is sRGB - there's no profile on the BBC web site image.) The "white" of the dress is also very far away from white (and very different from the white background of the page). Typical readings are R=135, G=135, B=175.
The colours of the dress in image on the Beeb web site were nowhere near black and white on an accurately rendered display. As the Wired article explained, it's quite possible to render an image that makes colours look different. It's also quite possible that the same image will look different in different lighting contexts, or even look different to different people in the same lighting context. But I'm quite sure that the primary issue here is a very badly coloured photo, and not people "misreading" the image, nor optical illusions.
-
But I'm quite sure that the primary issue here is a very badly coloured photo, and not people "misreading" the image, nor optical illusions.
There's no doubt it's a 'badly coloured photo' - the exposure is wrong, and the colour temperature is wrong. What's interesting is that most people don't see the dress as the colour numbers given by your color picker (blue/brown). Instead they see it as blue/black or white/gold. As we now know that the dress is in fact blue/black, surely those who see it as white/gold are misreading the image?
-
There seems to be a few factors that are contributing to the consternation revolving around this social experiment, but mostly it is about a lack of reference and a lack of a standardized, controlled experimental method for trying to detect and measure a phenomenon. With all of that ambiguity, what are we discussing other than a flawed process?
If anything, this points out the wide variation in people's interpretation of a supposed constant, known stimulus. Add to that the idea that the stimulus may not actually be remaining constant across viewers, as well as the bias and unreliably variable descriptions of the viewer's interpretation of the stimulus and it is no wonder the whole "discussion" is a confusing morass of disagreement. There is no control over the process or the variables requiring control to measure the signal.
Hey, at least it gave me the opportunity to bore the crap out of my kids by explaining and demonstrating to them the difference between incident and reflectance metering, and why black cats and snow are gray, unless you do something about it.
kirk
-
http://www.vice.com/en_uk/read/heres-whats-happening-with-that-dress-with-the-confusing-colors-882
-
There's no doubt it's a 'badly coloured photo' - the exposure is wrong, and the colour temperature is wrong. What's interesting is that most people don't see the dress as the colour numbers given by your color picker (blue/brown). Instead they see it as blue/black or white/gold. As we now know that the dress is in fact blue/black, surely those who see it as white/gold are misreading the image?
This just shows the extent to which color perception is contextual. Read the overall context in one way and you get white/gold. Read it another way and you get blue/black. There'd be far less hubbub about this if folks could come to terms with the fact that color is perceptual and not objective. Nothing has or is any particular color.
-Dave-
-
Sorry to resurrect an old topic, but it seemed better than starting a new one just to add an interesting link for anyone still interested. The latest issues of Wiley's Color research and application has an article that takes a true color science approach to the problem, which is refreshing after all the dumb media coverage. It's quite short and accessible. These are normally behind a paywall, but this one is open: http://onlinelibrary.wiley.com/doi/10.1002/col.21966/full
Also, it ends with a promise that the "Journal of Vision intends to publish a special issue on the perceived color of this dress."
-
Thanks Mark, very, very interesting!