[argyllcms] Re: Profiling with a known gamma shaper

Pascal de Bruijn wrote:
> Hi,
>
> I had another odd thought...
>
> In the past I've had two methods of profiling my camera:
>
> 1. Directly profile against the reference values of my IT8 target,
> giving good color accuracy...
> 2. Profile the difference between an in camera processed image and the
> IT8 target, giving me a great gamma shaper (very similar to camera
> generated JPEGs).
>
> One of the problems I'm having with the first method would be the fact
> that I suffer more often from blown out details...
>
> So I had a though, what if I just transplant the shaper from one
> profile to another. Which I tried with the color.org ICC Profile
> Inspector, which allows you to export and import gamma shapers, into
> and out of profiles.
>   

I think that I do not fully understand your objective, i.e. what result
do you exactly want to achieve? My feeling is that you basically do not
want to get a colorimetric transformation, but a perceptual one, which
emulates the perceptual transformation done by the camera when it
creates the JPEGs from the captured scene? But if this is the case,
what's wrong with your method (2) then? Or do you possibly only want to
assume the gray-axis mapping of the camera JPEG, but establish your own
mapping for non-neutral colors? But if the latter is the case, how do
you want to get the non-neutral colors mapped? Or do you have completely
different objectives?

Regards,
Gerhard

> The results were unsurprisingly not that great!
>
> The next obvious question would be, is there a way to generate a
> profile, while having a predefined gamma shaper?
>
> I'm aware this could entail losing color accuracy. But please when
> possible, indulge me...
>
> Regards,
> Pascal de Bruijn

Other related posts: