Paint of car has angle issue: different color in different view angle.
Multi-angle metering is needed IMHO.
Some car paints have angular dependence, not all. In any case, there is
a famous set of spectral reflectance data that includes Dupont car
paints. The article was
Vrhel, Michael J., Ron Gershon, and Lawrence S. Iwan, �Measurement and
analysis of object reflectance spectra,� Color Res. Appl. 19, 4-9, (1994).
In the past, the data were available by ftp from server ftp.eos.ncsu.edu
in directory /pub/spectra. I am not finding the files on that server
now. If you will contact me directly, I can send the whole set of data.
In the future, I could make them available through my web site.
The email address on my web site may or may not be working.
Jim Worthey
jworthey (at symbol) rcn.com
http://www.jimworthey.com
Metallic and pearlescent paint finishes can have
extreme angular dependence. Plain paint finishes
do not. But plain paint probably accounts for 20%
car finishes, at the most, so the statement that multi-
angle metering is needed is in general correct.
Alan
It seems that you are searching BRDF of car paint,
here is it,
http://www.graphics.cornell.edu/online/measurements/reflectance/carpaints/index.html