Re: Metamerism vs Color Inconstancy, again
Re: Metamerism vs Color Inconstancy, again
- Subject: Re: Metamerism vs Color Inconstancy, again
- From: Graeme Gill <email@hidden>
- Date: Tue, 22 Jun 2010 12:07:42 +1000
Marco Ugolini wrote:
Berns is too, apparently. Shouldn't we tell him?
I might chat to him about it if you like, next time I
see him at CIC.
Do we really need to make things more complicated than they need to be for the topic at
hand? Whatever else is tied to this concept, in the quote Berns talks about color
inconstancy IN THIS CONTEXT, not in that of discounting the illuminant and so forth.
You're the one dragging in a term that has much wider application that
metamerism. If you don't want to complicate it, don't bring Color Constancy
into it.
Can we please just stay on topic and not divert the topic, specially when this "wider"
meaning does NOT invalidate the meaning presented here?
It doesn't invalidate it, but it's kind of irrelevant if the underlying
interest is in the robustness of a metameric match, not the more general
phenomena of color constancy.
I really fail to grasp your meaning ("the reverse" of what?). Berns is saying that a
highly color-inconstant sample will tend to differ from a far more color-constant
reference sample more markedly upon a change in illuminant (the left-hand square - the
reference - remains grayish, whereas the right-hand square - the match/non-match -
turns pinkish under incandescent).
Right. If the tri-stimulus values change markedly due to a spectral change in
the illuminant, it's reasonable to expect that the color constancy
might be affected. But it is not a certainty, because color constancy
is the product of more than tri-stimulus values.
You may well have a situation in which a change in lighting induces large metameric
failures (compared to the target colors), but perceptually remains a constant color.
But that would be because the reference is color inconstant as well, wouldn't it,
whereas Berns makes the case of a (relatively) color-CONSTANT reference sample (from
one illuminant to another).
No, it might be because other visual phenomena dominate (ie. spatial color synthesis
mechanism, etc.)
But color inconstancy is NOT due to metameric failure. The latter is not its CAUSE:
metameric failures occur because one or both samples present a degree of color
inconstancy.
This would be a completely different meaning of the term "color constancy", and
the quote from Roy Berns doesn't imply this as far as I can see.
The accepted meaning of Color Constancy is the tendency of colors to remain
constant (when compared to our internal references, ie "Blue", "Red" etc.) with
changes in illuminant. Chromatic adaptation is one mechanism that partly explains this,
but it doesn't completely explain it. There are other mechanisms, such
as spatial comparisons, etc.
"Specific"? "Unrelated phenomena"? How is color inconstancy "unspecific" or
"unrelated"? Care to expand?
See above. It is the product of several different physical, visual and cognitive
mechanisms.
How does this work on color constancy negate or make invalid the meaning of color
INconstancy explained by Berns? (...which is the meaning that I have been trying to
expound upon, against rationally unexplainable opposition on this forum, seemingly more
interested in finding reasons to counter it, however elaborately strange these reasons
may be, rather than in conceding this quite uncomplicated point of distinguishing
conceptually between "metamerism" and "color inconstancy".)
It doesn't. But I suspect you've drawn unwarranted conclusions from Roy Berns
note that changes in tri-stimulus values due to spectral interactions
can have an impact on color constancy.
Graeme Gill.
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden