I have been trying to re-use displacement maps from one character to another.
The goal here is to work once, and take avantage of your effort many times.
I will try to make myself as clear as possible. I have one character in T-Pose (we´ll call it Character A) and
then when have the same character in a different pose (we’ll call it Character B). Well,
they are exactly the same mesh, except for the pose. Same point order, polygon count, polygroups,
UVs, everything is the same, except, again, for the pose. It is a single mesh, single group, subtool-less mesh.
I pretend to use the T-Pose character to create a nice displacement map, after a lot of hard work, to create sort of a “master character”
that will drive the mesh details of any other pose that I happen to create based on that same
character. You can easily do this in Poser, provided that you export with the correct options.
So, again, create a master character in any application (I happen to use Poser) export into Zbrush,
work up the detail in higher subdivisions, finalize your detailing, create a displacement map to lock the
details from subdiv 1 to subdiv 5 (around 790000 polys in my case), save everything and go to render.
Well, I have the displacement map for character A. This displacement map drives the geometry of character A
from the intial 29500 polygons to say around 790000 polygons. I am using Maya and Mental Ray for the coupling of the
subdiv 1 mesh and its displacement map. So far, so good.
Character A, in T-Pose, renders perfectly, after a little dooling, in Maya. I am happy with it so let´s go on.
I now take Character B (I shouldn´t be calling them Character A and B, but rather Pose A, Pose B, sorry!) with exactly
the same polygons and vertex count/order than Subdiv 1 of Character A but posed differently. I go to Maya and apply the same shader network
that I used with character A, because they are exactly the same mesh. The displacement Map can be used for both poses
because we are talking about the same UV map here. Also, the subdivision approximation that Mental Ray uses for the
mesh uprezing prior to the application of the displacement map itself is the same here, because we start with a 29500 polygons
mesh and we want to end up with a 790000 polygons mesh before applying the displacement map in Mental Ray.
You hit the render button…and…sxxt!
You can “feel” that the displacement map its doing its job, however, there is a problem here.
When you work on Character A in Zbrush, as we all know, the changes that you apply in higher subdivisions do get
propagated down to lower ones. That is, subdivision 1 of character A gets a little tweaking from higher subdivisions.
This changes, although subtle, in subdivision 1 of Character A, make the mesh “different” from Character B. Since
displacement, as the word implies, displaces the mesh (in the normal vector direction) during rendering from the starting mesh it was created from, the
differences between both subdivision 1 (Character A) and Character B make the final result quite different.
Character B is say “a clean mesh” out of Poser, while Character A´s subdiv 1 is a tweaked mesh in Zbrush. The displacement applied
to both characters, although the same, does not give the results expected.
One could think…well, pose the character in Zbrush and problem solved, however this is not an option in my case, because
Poser makes your life in this regard a whole easier.
So here is my question. Is there any way to copy the subtle changes in subvision 1 of Character A to Character B before
applying the displacement? I tried everything my feeble mind could think of… projecting, importing one mesh on top of the other,
reconstructing subdivisions, etc… with no sucess.
Can anyone help?
Thanks
Tomas