Re: Rendering Intents
Re: Rendering Intents
- Subject: Re: Rendering Intents
- From: Graeme Gill <email@hidden>
- Date: Sat, 31 Aug 2002 18:46:53 +1000
michael shaffer wrote:
>
This would beg the question as to "why?" That is, ... a conversion
>
between gamuts of similar size would ask for very little compression ... so
>
why does perceptual fail? Are there more fundamental differences 'tween the
>
2 intents?
I can think of two reasons that may explain this.
One is due to a fundamental limitations of the ICC profile
design, that the gamut mapping is implemented primarily in
the output table (B2A table), without any knowledge of what the source
gamut actually is. I would therefore presume that a perceptual
B2A table in most profiles, will compresses regardless, thereby
producing an inferior result when no compression is actually needed.
A sensible gamut compression scheme does so with the full
knowledge of the source and destination gamuts, and (if it
worked well), one might expect a perceptual and colorimetric
intent to produce very similar results if the two gamuts
are similar.
A second reason may be that many of the popular perceptual
gamut mapping algorithms use simple geometric constructs to map
colors, and these may distort the mapping, even if the source
and destination gamuts coincide.
Graeme Gill.
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.