Re: fine art reproduction questions
Re: fine art reproduction questions
- Subject: Re: fine art reproduction questions
- From: "Tim Brown" <email@hidden>
- Date: Mon, 3 May 2010 10:44:00 -0700
- Thread-topic: fine art reproduction questions
Use of spectral data, system data, light fall-off is really cool. It
also beats the alternative method of trying to line up the lights to be
somewhat even, and using a cmyk printed chart to calibrate a camera to
measure non-CMYK artwork.
When capturing original art, one often has the intent of capturing the
texture, as well. This is usually done by a bit of side-lighting, which
causing additional difficulty in making the amount of light uniform.
Another difficulty includes the reflection of the illumination from the
texture to the camera.
All told, these cause significant issues to the spectral correction
system. Specifically, polarizing filters are needed to cross polarize
and cancel out the specular highlights. Even when these are measured as
part of the capture system, these negate the ability of the system to
handle the color shift.
Using this system for many years, we have noticed that the hotspot
correction works great. The general color correction is consistently
off a bit - heavy in the yellow/green regime. Also, the amount of light
needed when cross polarizing is quite substantial. We typically produce
a lot of heat, and have warped (and replaced) many illuminator filters
and lenses.
As for touching the surface, the options fall into 3 categories.
1 - Don't care -- living artists, high-volume captures; go ahead, touch
it with the spectro.
2 - Known materials on the untouchable canvas -- if you are working
through a museum collection, you have a pretty good idea of the
materials used by type/period of art; build yourself a collection of
similar colors on period canvas. Most of the colors will be fairly true
for the type/period of painting. You can't touch the art, but you can
scan some similar colors made from period paints on period canvas.
3 - Unknown materials on the untouchable canvas -- don't know if the
artist used real or synthetic ultramarine? Vermillion or
vermillion-like? In this instance, use your period approximation and
edit by eye.
The correction isn't going to be perfect - this is still an
approximation...the best approximation possible, using the most data,
but still an approximation.
My 1.5 cents...
Tim Brown
Senior Engineer
The Thomas Kinkade Company
_______________________________________________
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