Thank you guys!
I took the liberty to copy this post I just made on zbrushcentral in which I elloborate a bit more about the progress. For those who are interested!
About your questions concerning the deformation and process about creating these faces. I aproached this model from a film production point of view. Meaning that the model doesn’t only has to work as still imagery but also will need the proper topological support to be able to move and deform correctly in animation.
Forming wrinkles in a polygon mesh gives you a completely different effect then blending in wrinkles in a painted or extracted displacement map. The effect of flesh and form compressing into eachother I believe can only be achieved properly in a realistic manner when it’s actually modeled in. The displacement detail is just along for the ride.
The basemesh I build therefor is fairly dense. All the big wrinkles and most of the wrinkles around the eyes were modeled into the original model in Maya. Most of the deformation for the expressions takes place there. The displacement detail I sculpted in Zbrush was applied to these shapes and corrected appropriatly later. These so called Driven Displacement Maps can be hooked up in the shader tree and connected to the blendshapes within Maya so they dial in when nescecary. These corrective displacement details were fairly minor, but absolutely crucial to achieving a realistic effect on screen. I’ve attached an image to show you an example of the corrective displacement work I’ve done.
The ‘Rage’ mouth on the left shows you the original displacement detail applied to the blendshape without any adjustments made. The ‘Rage’ mouth on the right shows the corrective work I’ve done, fading out the wrinkle detail on the lip, which happens when the mouth is pulled open this wide. The lips stretch around a much larger area and loose the crumbly lip wrinkles that showing in more neutral expressions.
The other image I’ve attached shows you the original Maya model and the ‘scared’ blendshape. Here you can see what happens in Maya and what was adjusted in Zbrush if you compare it to the final render.
I realise that for the goal of creating 15 still images, this approach might seem overdone or unnescary. I could have done it all in Zbrush, but in the end it saved me time. Ofcourse if I would ever have the intend to get this animated, it would actually work, which is a nice idea.
If you want to know more about driven displacement maps I can suggest to pick up a copy of Hyper-Realistic Creature Creation which is beeing published by Autodesk and is co-authored by Jeff Unay, one of the facial leads over at Weta Digital.
Hope you found this useful, let me know if there’s any other questions!

