PDA

View Full Version : Morphs, couldn't they be interpolated by edge lenght?



probiner
12-11-2010, 01:20 AM
Back to LW and crazy ideas.

A morph and a bone chain animated A into B.
90781

So the general shape in A and B is the same, only some edges changed lenght and angle.
90780


Question is: Wouldn't there be a way that morphs would be more than A to B, but use the different edge length (and maybe angle) from A and B to interpolate the morph deformation?
I have no idea of what this involves, but in days where we see lots of things coming out of nowhere, things seem simply possible.

90779
^ Morph (Green) and "edge" changing angle and lenght (Red and Blue)


From the top of my head, i guess the issue would be to find pivots and cache for so much data...

Cheers

probiner
09-08-2012, 02:53 PM
I think I might have explained myself wrongly in my first post, especially with the last image.
This hit me again today but more like this:

- Morphs using Dynamics Calculations. Position A, Position B for the mesh vertices and let the dynamics solve the in between with some iterations back and forward, to get a good unfolding of the perceived volume from one shape to the other, instead of linear morph. Does this make sense?

Cheers