gdalgorithms-list

Re: [Algorithms] RGB to XYZ conversion

[Mat N. <mat...@bu...>]

He missed a few signs:

> [y,Co,Cg]' = [1/4, 1/2, 1/4].[r,g,b]'
>              [1/2, 0,   -1/2]
>              [-1/4, 1/2, -1/4]

MSN
-----Original Message-----
From: gda...@li... [mailto:gda...@li...] On Behalf Of Jon Watte
Sent: Wednesday, April 23, 2008 10:58 AM
To: Game Development Algorithms
Subject: Re: [Algorithms] RGB to XYZ conversion

Sam Martin wrote:
> We recall from [1] that the direct YCoCg color space transform is defined by:
> [y,Co,Cg]' = [1/4, 1/2, 1/4].[r,g,b]'
>              [1/2, 0,   1/2]
>              [1/4, 1/2, 1/4]
>

That can't be right, as it collapses RGB into equal values for y and Cg.
You can't possibly reconstruct RGB from that.

sincerely,

jw



-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
GDAlgorithms-list mailing list
GDA...@li...
https://lists.sourceforge.net/lists/listinfo/gdalgorithms-list
Archives:
http://sourceforge.net/mailarchive/forum.php?forum_name=gdalgorithms-list

Re: [Algorithms] RGB to XYZ conversion

[Tom P. <ga...@fa...>]

> Dan Glastonbury wrote:
>>  What matrix should I being using to convert RGB values to CIE
>> XYZ?
>
> Good question Dan, and I for one don't know the answer.

I think the problem comes down to the fact that XYZ is an absolute
color space where RGB is flexible given the settings of your display
hardware.  Therefore the same RGB values piped through different
equipment would yield different values for XYZ if sampled with
suitable equipment.

This may be less of an issue if you're using XYZ as an intermediate
format, though, since presumably you're going to start and end with
RGB, and are doing work in XYZ space to have your transforms work
more "naturally" or whatever.  If this is the case, anything that's
out there is probably sufficient for the task, as long as you can
cleanly invert the operation for all of the values you're looking to
use.  Mind, there will be colors in XYZ that won't be representable
in RGB.  At least, as I understand it...

Sorry, I don't have any citations; I was researching this some
number of years ago for some color tools that I was working on, and
basically came to the understanding that as RGB is not an "absolute"
color scale, not to mention it is very limited, I just needed to
define the space where my operations "made sense" and transform into
and out of that space.

-tom!

Re: [Algorithms] RGB to XYZ conversion

[Dan G. <dan...@gm...>]

>  Good question Dan, and I for one don't know the answer. I asked myself
>  the same question when I wrote that blog post and first worked out the
>  code.

When going RGB->LogLuv->RGB I don't think it matters too much.  I'm
not expert on these matters, but it's foreseeable that the conversion
could effect the range of x & y and thus the "magic" constants.

When converting between different colour spaces then the "correct"
matrices would be important.

cheers
DanG
-- 
Dan Glastonbury, Dan dot Glastonbury at gmail dot com
`Pour encourjay lays ortras'