white balance of human eye

[Chris Burke]

What is it considered? I was shooting under tungsten light and of course when I white balance to a white card, the overall balance is neutral, white is white. However the color in camera and on screen is not how I perceive it. I see a warm cast under tungsten light. So it got me wondering, what is accurate? How is color  perceived by us humans, all things being equal. I realize that "look" varies from project to project, but what is considered true? I would think what we see and how we see it is normal, not what a sensor says it is. I shoot mostly on film and Kodak or Fuji when I shot it, rendered color much closer to how I see it. 

[Max Field]

Before anyone makes any claims on what their eyes perceive, it's important to take one of these tests first:

https://www.xrite.com/hue-test

As for my own experience, your brain tricks your eyes into a nearly automatic white balance. We get used to certain tones and when a new one is introduced we sort of slowly get accustomed to it. When I first saw tungsten light from a real fixture I thought it was insanely yellow. For a year I had a 2200K bulb in my bedroom, then when I flipped on an MR head, it seemed pretty white to me.

I guess we have color meters for a reason idk.

[Phil Connolly]

I wouldn't be surprised if from an evolutionary perspective Humans eyes work best under daylight balance. That would have been the main light source for the duration of evolution.

But it seems we can't tell obsolete colour because our brains interpret the images. 

As our eyes age, I've heard the lenses in the eye go more yellow creating a yellow colour cast that the brain probably corrects out.

[M Joel W]

 

4 hours ago, Max Field said:

Before anyone makes any claims on what their eyes perceive, it's important to take one of these tests first:

https://www.xrite.com/hue-test

As for my own experience, your brain tricks your eyes into a nearly automatic white balance. We get used to certain tones and when a new one is introduced we sort of slowly get accustomed to it. When I first saw tungsten light from a real fixture I thought it was insanely yellow. For a year I had a 2200K bulb in my bedroom, then when I flipped on an MR head, it seemed pretty white to me.

I guess we have color meters for a reason idk.

I scored a two, with a weakness in greens. So I might not be one to talk as I don't have perfect color vision, but I once tried something really odd, staring outside with one eye while staring at a flashlight with another. Something along those lines. Anyway I think each eye adjusted independently because one eye had a warmer tint than the other until the two grew acclimated. Kind of like those tricks where you see an after image briefly after staring at an inverted image for a while.

Clearly I was bored. I suspect the eyes do a lot more local adaptation than we realize:

https://en.wikipedia.org/wiki/Checker_shadow_illusion

 

Edited March 2, 2020 by M Joel W

[David Mullen ASC]

I scored a zero, which it says is perfect. Perhaps it's not something that declines with age...

[Phil Rhodes]

Lenses in eyes do go yellow as part of the process of getting age-related cataracts. I've spoken to people who had lens replacements done for that reason in one eye only and described the per-eye difference as a light straw colour. Generally the visual system will balance it out.

The human visual system is capable of similar colour balance errors to a camera. Sometimes this is visible when working on a greenscreen stage. Stare at the screen for ten seconds, ensuring it fills your whole field of view, then turn around and look out the door; the world flashes magenta for a moment. This can be simulated on a computer screen in a dark room, but there's the factor of the screen filling peripheral vision and removing all other colour references that makes it work well.

Humans are fairly good at comparing two colours they can see simultaneously, which is why it's really important that lights we use on set all match.

Humans are very bad at comparing colours they've seen consecutively. A test was run in which people were shown a test colour patch; the patch was removed, then replaced with several similar patches of which one was an exact match. Most people could find the matching patch after a delay of up to... four seconds. Honestly, it was four seconds. Human colour memory is hilariously bad.

This is why I suspect that all this obsessive monitor calibration that goes on is probably more to do with people's need to pretend they have some sort of superhuman golden-eyes capability than any real technical necessity. In terms of colour, monitors could be much more wrong than they generally are, and nobody would have the faintest idea unless there were two displays side by side. That's not so much the case in terms of brightness, where potentially-important shadow detail is either visible or it's not, but as far as colour goes, nobody (and I do mean nobody) can tell from session to session whether something is a handful of delta-E off or not. This fact is regularly leveraged by low-budget movies finishing in poorly- or un-calibrated environments. You might not be able to control precisely what it will look like, but you can achieve something that's entirely acceptable. Nobody can tell. It's fine. Sorry, but it is.

And the idea that there's a need for tightly calibrated on-set displays for different people of different ages with different eyes, having just wandered in from places with different lighting conditions, in a completely uncontrolled environment, is pure comedy.

P

[Stephen Perera]

Fun test...I scored a 0 - perfect score, on the x-rite test do i get a badge? hahah