I think perhaps there's no reason one has to stick to a static construct with
the only change being the camera location.

Since for every animation frame the scene can change, perhaps the way to do it
in one shot is to dynamically adapt the media density to what that frame
requires to achieve the desired look.

So graph out what you want in terms of visibility, and then based on the
distance between the camera and the planet's surface, change the density to fit
your needs at that instant.

Similar to what others have proposed, you could define your visibility as a
spline, interpolate, and use the camera distance to the planet to index a
density value.

There's probably other ways to do it, but I'd say the first step would be to
story-board out the animation and use that as a guide to define very detailed
requirements for key points in the animation, and then see what the best method
to transition between those points is likely to be in terms of time, complexity,
and quality of end result.

Just my take on how I'd likely approach this.

Post a reply to this message