RE: relative colorimetric without black point compensation
RE: relative colorimetric without black point compensation
- Subject: RE: relative colorimetric without black point compensation
- From: Roger <email@hidden>
- Date: Sun, 25 Jan 2009 11:32:39 -0500
Graeme,
Let me push this point just a little bit further.
Correct me if I am wrong but I tend to think that, although I never got
proof of this idea, in coming up with "true" perceptual mappings to the PCS,
any profiler has to make some kind of assumptions about the Source (some
Source).
I know Argyll is slowing getting its feet wet in this area, of creating a
true, unique Perceptual rendering intent, differentiated from RelCol, and
you probably read dozens of articles on how this is done in the literature.
Wouldn't that be a necessary assumption, though, in your opinion, that an
astute profiler would have to make this?
It is a fact that many profilers don't bother to populate the Perc tag with
real Perc data while a handful of profilers take the pain of conscientiously
build meaningful Perc RI. Some, even having very good invertibility, except
at the very edge of the gamut, where it is said to be nearly impossible.
Which leads me into PRGM. Isn't that saying a lot about the Source?
Best / Roger
> Given that perceptual (typically) knows nothing about the source, it
> doesn't
> take much. BPC is computed on the fly from the source dynamic
> range and destination dynamic range. Of course it does nothing
> about any mismatch in the a* and b* directions.
>
> 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