Dr. Renzo Shamey Explains the Science Behind #TheDress

You’ve probably seen it by now – the photo (below) of “the dress” making the rounds on the internet that people just cannot agree on. What color it is? Is it white and gold or is it blue and black? And how can a dress appear so differently to different people?

Internationally renowned color scientist and professor in the Wilson College of Textiles, Dr. Renzo Shamey, weighed in with his opinion of just what is going on with #TheDress on March 2, 2015, live on Good Day DC (Fox). See the broadcast here. Or read below for more!

When we look at an object in a scene we are influenced by the surround and our visual system is also influenced by the intensity of the illumination (as well as its color).  I have seen three different versions of this picture and they look markedly different (to me!).  This shows problems surrounding digital color communication and why it can be so important. The number of people disagreeing with one another and the level of emotion surrounding the topic also shows that we really care about our colors!

So before I go on further let’s remind ourselves that the perception of color actually happens in the brain and that signals obtained from a scene are interpreted in our brain and that interpretation is influenced by past experience and also the type of information present. Cones responsible for color vision operate at higher levels of illumination and generate signals that are then interpreted as Red, Green, Blue, Yellow, etc.  It is well known, however, that our perception is strongly influenced by surround.  Darker surround can induce the stimulus (in this case dress) to appear lighter and vice versa.  We also use cues from our surround to maintain color constancy.  That is, when we look at a scene that is partially in shadow we process, internally, the colors such that we know they are the same, whether they are in shadow or not.  If we don’t get sufficient cues from the image that we are looking at, such as skin color, sky, white board, green grass, etc. then we may be confused when we are processing the limited information that is available in the scene.

In addition, the difference between some colors, e.g. gold, orange, brown is mostly dependent on their variation in lightness.  In fact it can be shown that these will appear the same to an observer under certain conditions.

So I guess you are saying enough of this just tell me what the color is…. In one picture I saw the dress as white (bluish white) and gold (with black edges).  In another pic I saw the dress as clearly blue and black and yet in the third picture I saw a combination.  This shows how difficult color communication can be, if proper precautions are not taken into consideration. Also, when I looked at the top of the dress and then down, my perception was influenced (or I could tell why it might be influenced) by what I saw as the background color of the dress (in this case dull orange) whereas when I looked at the lower section of the dress and then up I could see that blue and black became a lot more dominant.

Incidentally the actual color of the dress is apparently blue and black, yet a change in illumination, angle of viewing, level of illumination, surround and each observer’s level of color and light adaptation will influence what they see! At the end of the day you see what you see and we still do not fully understand how color percepts are fully formed.

Take also into account that the image can be white balanced (calibrated) differently and this will shift colors in certain directions depending on the color of the source used.

But why do two individuals looking at the same picture see it differently? Part of the reason may be due to variations in their adaptation (both chromatic and light) and also the cues their internal visual processing system uses to come up with the most plausible color for what they are looking at.  That is, if they are both color normal, and do not suffer from any form of color deficiency, if they are able to describe colors more or less in the same fashion and if they are using the same monitor (or display device), same angle of viewing and are of more or less similar age.  You can guess that all of these factors, e.g. viewing and illumination conditions, age, gender, etc. may affect perception and they indeed do!

There are a lot of other sources of variability among people, such as the distribution of photoreceptors in their visual apparatus, and the way they process color internally…our understanding of the visual system has improved vastly over the last several decades but there is still a large section of the processing that in the dark so to speak!

Isn’t color fun?!!!