[Openicc] Re: color-policy vs. color-infrastructure

[Maria, Marti]

Hi,

> Gamut tags are a bit fuzzy at the best of times, and in many profiles 
> out there they are simply wrong. As a result, they can't generally be 
> relied upon.

Just for the record, it seems Photoshop is not using the gamut tag
anymore. Lcms is actually using a brain-dead method for inferring gamut 
that works most of times at expense of poor accuracy. The trick is to
convert values back and forth the colorimetric intent. If color changes,

then it is outside gamut. 

I'm right now investigating more advanced methods to do that, and high
in 
my priority list are segment maxima, mountain range and finally alpha
shapes, 
but this latter seems to be too slow and complex for real world usage. 
Convex hull is not suitable at all, since many gamut hulls are
non-convex.

Regards,
Marti Maria.


-----Original Message-----


Kai-Uwe Behrmann wrote:

> Could argyll and lcms provide a mechanism to take a profile as an 
> argument , and give back either the gamt tag or the gamt tag added to 
> the former profile?

Yes, that's certainly possible.

> As Marti pointed out on his list gamut tags may be not that reliable 
> for usage. And I have the impression too, especially for older
profiles.

Gamut tags are a bit fuzzy at the best of times, and in many profiles
out there they are simply wrong. As a result, they can't generally be
relied upon.

> Thats maybe an information applications can easily provide by 
> intelligent scaling down an small sample image and using the ncl2 tag 
> to store its colours (I do in CinePaint).

A hierarchical image format would certainly make creating such a gamut a
lot faster.

> I am highly interessted in this toppic. As it seems the basis to 
> sucessfully use CIE*Lab as editing and transportation mode for image 
> content.

Determining and transporting gamuts does seem to be part of the puzzle.

Graeme Gill.