Search the Community
Showing results for tags 'facial rigging'.
Found 2 results
Hi everyone, I'm usualy a maya user and I'm starting to use Houdini at a personal level for the moment. My goal is to try to do a complete character facial Rigg and I'm gonna need some help I think. I usualy do a combination bones/skinning and blendshapes. I'm wondering how to mix different shapes and apply a corrective blendshape on a combination of 2, 3, 4 blendshapes for example. I saw how to connect basic blendshapes, the problem is how to call a third blendshape when you activate two shapes ? And is there a tool that enables you to sculpt a mix and extract the corrective blendshapes? Do I have to do this manually but substracting the shapes on my mix to get the corrective shape and make an expression that activate blendshape C(corrective) when mixing blendshape A and B ? The way I use to do that was by using a plugin call BCS for maya. It enables you to easily split and mix several blendshape and that what I'm looking for in Houdini. It may not be the right philosophy to try to do this way so If anyone have some experiences in this area or some advice that would be very much appreciated. Have a great day and hope to get some feedback. $ Cheers,
So i thought i might post something i have been working on lately. It was from an old file i never finished, and thought i might turn it into something. Textures are hand painted (in 3dcoat) and its just the starting pass (no details yet). I'm pushing my system and Houdini to see how it handles (8 x 4K Textures). The model is a straight import from Zbrush at the moment which is around a 750 polys (which has little no details). The final mesh is around 21 million polys, which might seem excessive but you will see the details that are in the model soon enough. Still a lot of painting left to do but i need to see how they are reading in the render. SSS maps are generated from the base maps and point cloud about 200,000 points for accurate coverage though i might try higher. Render times are fairly good. About a 90-110 seconds at the moment. But there is no spec, reflection, disp, or density maps. In terms of tools, just standad H13, made a custom textureVOP and colorVOP so i can plug in multiple maps faster (turn them on or off etc), working on a procedural eye generator (still early days but the geometry is almost finished). I have done some facial rigging tests and for what i want, i will need to make a custom blendshapeVOP but i have a new (artistic driven) blendshape method that is pretty good at simulating skin sliding which is nice. Lots more on the way, this is a wip Here is the starting point. (Beauty render + SSS contribution) Looks like the sss needs to be stronger?