bertjenkins
07-29-2010, 08:44 PM
Hi,
I'm looking for advice on workflows for hi rez facial animation. I'm working on a chimpanzee project and i have sculpted the animal's head in mudbox. I'm going to rig and animate in maya and i want as much resolution in terms of facial deformation as is workable.I spent ages getting the topology right as i really want the skin folds/ creases and muscle groups to feel like they are deforming realistically. However before i start rigging i have a few questions as I've not done anything as hi rez/ photoreal as this before.
Has anyone got experience using vector displacement maps over a low res mesh? Is this worth pursuing? Are the deformations convincing? (ie the wrinkles etc) Are they tricky in mental ray?
Is it more reliable just to use a high res mesh with a wrap deformer?
Or is there an alternative workflow for the lo res/hi res proxy method?
Should i expect to be retopologizing, now that the high res mesh has new wrinkles, (bags under eyes etc) even though the base mesh has pretty good topology?
Also how do i go about creating the muscular stretches as seen on softimage's face robot, etc? Can i achieve this using blendshapes/ clusters alone (I'm familiar with Osipa, etc) ? Or are there any better methods to get deatail in the muscles, build FACS poses etc?
Thanks in advance, if you can answer any one of these i'd really appreciate it. Just want to avoid going all the way through rigging and skinning and realising (when it's too late) there's a much better way of going about it.
I'm looking for advice on workflows for hi rez facial animation. I'm working on a chimpanzee project and i have sculpted the animal's head in mudbox. I'm going to rig and animate in maya and i want as much resolution in terms of facial deformation as is workable.I spent ages getting the topology right as i really want the skin folds/ creases and muscle groups to feel like they are deforming realistically. However before i start rigging i have a few questions as I've not done anything as hi rez/ photoreal as this before.
Has anyone got experience using vector displacement maps over a low res mesh? Is this worth pursuing? Are the deformations convincing? (ie the wrinkles etc) Are they tricky in mental ray?
Is it more reliable just to use a high res mesh with a wrap deformer?
Or is there an alternative workflow for the lo res/hi res proxy method?
Should i expect to be retopologizing, now that the high res mesh has new wrinkles, (bags under eyes etc) even though the base mesh has pretty good topology?
Also how do i go about creating the muscular stretches as seen on softimage's face robot, etc? Can i achieve this using blendshapes/ clusters alone (I'm familiar with Osipa, etc) ? Or are there any better methods to get deatail in the muscles, build FACS poses etc?
Thanks in advance, if you can answer any one of these i'd really appreciate it. Just want to avoid going all the way through rigging and skinning and realising (when it's too late) there's a much better way of going about it.
