Re: Camera Profiling question
Re: Camera Profiling question
- Subject: Re: Camera Profiling question
- From: Graeme Gill <email@hidden>
- Date: Mon, 15 Sep 2008 00:42:00 +1000
Richard Apollo wrote:
The Star-Bellied Sneetches want the elements in the photograph to render the same color
under all lighting conditions, therefore camera profiling doesn't work.
The two aren't actually connected. A relative colorimetric intent profile
makes the profile white point relative, hence performing a white balance,
but the underlying absolute characterization of the device is white
balance agnostic. Using an absolute colorimetric profile before
white balancing should be a perfectly viable approach, degraded
only by the degree of spectral dissimilarity between the calibration
and actual scene illumination (and I'm not meaning white point
difference of the illuminants here, I'm talking about the interaction
of the illuminants and colorants spectrums). After all,
this is what a relative colorimetric profile is doing, albeit
with a fixed white balance.
Doing things the other way around (setting a white balance
before the profile) probably isn't the best thing to do
if the profile is meant to be compensating for channel
non-linearity, since (at it's most basic) a white
balance is set by changing the relative channel gains,
and changing the pre-profile channel gain will shift
the location of non-linearities, invalidating the profile.
In practice, if the sensor is reasonably linear, and
the RAW values are linear light, then I'd imagine it's probably
not going to make much difference whether the white balance
is set pre or post profile (assuming the black point offset
is taken care of in the RAW processing, and the profile is being
used to transform into an RGB like working space).
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