Well, I suppose I should explain the situation without a deformer plugin: I'm trying out Jason Osipa's Facial setup, in which he takes the head mesh and uses a wrap deformer to drive lattices that deform the eyes, teeth, etc to follow the mesh. Well my wrap deformations weren't looking so hot for which Jason recommends corrective shapes but I'm not good a modeling so that would probably make it worse. So my idea was to attach follicles on the head that were close to the lattice and then use the follicles transformation matrix to drive a skinCluster deforming the lattice. Only follicles take a lot of time to set up and don't offer any control over rotation order. Using a riveted transform would have been even more complicated. So now I'm considering just taking the mesh itself and treat it's components as transformation matrices to create a vertex/normal driven skinCluster. I guess a simpler way is to create a compound attribute which specify the indicies for the deformer to use. Now this brings up another question, how do you create a deformer that works on lattices? Can you pass a lattice MObject to MItGeometry?
On Aug 15, 5:17 pm, "Marco D'Ambros" <[email protected]> wrote: > How Chad says there is a different way to do it. > > yes in your case maybe I'll use a deformer, you need to decide you is the > deformer and how is deformed... > > but which closetPoint function should I use? > > everythong depend on what you want to do, for example I used a ray > intersaction ( because it's faster ;) and in my case the was not a > difference with the closest ) > > And what if there are multiple matrices matching a certain distance to an > array of > points? > > the same depend what you are looking for, but in theory it's just a > condition on the closest point, so just an loop on the multi point to > determinate the closest point. > > I hope that I'm usefull :) > this is a little example of one deformer that I made > > http://vimeo.com/26828121 > pw: test_01 > > -------------------------------- > Marco D'Ambros > phone : (+61) 0431 830085 > web :www.marcodambros.com > mail : [email protected] > > > > > > > > On Tue, Aug 16, 2011 at 5:14 AM, Viktoras <[email protected]> wrote: > > > On 2011.08.10 20:09, notanymike wrote: > > >> It's not a new idea to deform a mesh using a transform riveted to that > >> mesh. However I was wondering if one could do the same without > >> transforms. That is to say, something that's a cross between a skin > >> cluster and a wrap deformer. I was thinking of using MItMeshFaceVertex > >> to build a transformation matrix using the position as the translation > >> value, and the normal and tangent for rotation. The question is then, > >> now that I have this array of matrices, how then do I make sure they > >> deform the right regions of the target mesh? > > > Am I under right impression that you're trying to implement something like > > control points over a deforming geometry? e.g., you've got head with blend > > shapes, then skinned to head/neck, and then on top of that deformation you > > want something additive like "raise brow corner" transform? > > > -- > > view > > archives:http://groups.google.com/**group/python_inside_maya<http://groups.google.com/group/python_inside_maya> > > change your subscription settings:http://groups.google.com/** > > group/python_inside_maya/**subscribe<http://groups.google.com/group/python_inside_maya/subscribe> -- view archives: http://groups.google.com/group/python_inside_maya change your subscription settings: http://groups.google.com/group/python_inside_maya/subscribe
