[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re;Re Re Layering vrs Windows


I apologise if my meanings are unclear. In the first
instance I do not wish just to keep adding masks or
windows unless they can be made to be active, ie
contain a unique output, rather than just passive,
ie used to reinstate the original colour of either the
background or an object. Even then there is a practical
limit to how much correction a scene can take. If I had
to use every channel of DCP on every scene I came to 
there would be serious problems with either my 
methods or eyesight!

Re your second point;  yes of course a set of pixels
of a particular level of hue, saturation and luminance
which define an object are self tracking right now.
The problem comes when those same values of 
colour, (H,S,L), are present in other areas of the same
scene. At present we use masks,regions or windows
to stop interaction and then have to keyframe them
through the scene. I would rather have a computer
work out the shape of my chosen group of pixels
and track that shape for me.

Your last point; do I see much difference? 
No, not really. The two methods flow neatly
together in the system that I use and there is an obvious
logic to their use, ie, use hue, saturation and luminance
to get as much separation as possible and then if needed
use a region as another barrier. Meanwhile keep it all

I hope this makes things clear,

Seamus O'Kane
VTR Ltd. London.

thanks to Sam Dlugach, Dean Humphus, & Clark Bierbaum 
for support of the TIG in 1997
mailinglist digest available......posting guidelines on the webpage