Re: CM challenge for an architectural photographer
Re: CM challenge for an architectural photographer
- Subject: Re: CM challenge for an architectural photographer
- From: Roger Breton <email@hidden>
- Date: Fri, 10 Dec 2004 11:44:29 -0500
In the past, I have successfully shot products which I've adjusted in
Photoshop by dialing in the Lab values of the products, take with my Digital
Swatchbook. It worked very well then. I don't see why you could not do the
same things with walls and other architectural details? After all, what is
being reproduced through the film and your scanner profile is the way the
artitecture interact with available light as seen by a Standard Observer.
> A photographer asked me today how to solve a problem he's run into a
> few times when shooting interiors for architects. Here's the situation:
> he shoots 4x5 film, scans it and then does any necessary retouching
> before printing the image on an Epson 2200.
OK.
> The client wants the colors
> on the prints to match the colors of the actual walls etc. in the room
> he's shooting.
Matching colors colorimetrically and perceptually is not the same. The CIE
system and the ICC only takes into account the reflective or emissive
properties of a surface but say nothing about the conditions in which these
observations are made, which is the domain of color appearance models. The
commercial implementations of those are probably still a few years away.
Advanced digital cameras are starting to deal with this effect. Film has
dealt with that effect for a long time. But getting it into Photoshop-kind
of imaging system is another story. You may want to start by studying Kodak
RIMM and ROMM models, scene-refered colorimetry. There are excellent papers
on this on the net a few google search away. Papers and books by Mark
Fairchild from the Munsell Color Science Lab in Rochester, NY, is another
place to look.
> He was told that he could use the EyeOne to take a reading of the
> important colors in the scene and then use those readings to adjust the
> values in the images,
Correct. Although a non-contact instrument could be used for this as well
but it carries issues of calibration I have not resolved yet. But it's
feasible.
> but I see many variables in this idea that would
> seem to make this a very complicated solution. Sure he could know what
> the measured values are, but that wouldn't account for the lighting
> etc. in the room would it?
Nothing. If you only measure the reflectance of the walls with a 45/0
instrument, then it is saying nothing of available light.
> (for instance the EyeOne uses it's own light
> source and wouldn't know how the light in the room was affecting the
> way things looked in context.)
Any spectros, like the EyeOne, measurements are independent of ambient
light. And don't have to.
> My thought is if he has a profile for the scanner and for his
> printer/paper combination, his monitor is calibrated/profiled, and if
> his transparency is "right" he should be able to match colors without
> any major headaches.
Fair.
> Does this seem right, or would he need to go the
> more complicated route of measuring all the important colors and
> comparing them to the values in his digital files?
It would not hurt to validate his results with real world measurements.
> Or is there a better way any of you can suggest?
I think straight digital shooting could be better. Because the film
interpretative stage is skipped. And it's just scene luminance -> computer
digital counts.
> Thanks!
>
> Dennis Dunbar
Regards,
Roger Breton | Laval, Canada | email@hidden
http://pages.infinit.net/graxx
_______________________________________________
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