Speech Graphics
Bust – (September 2011)
(Click the images below to enlarge them, and zoom using the mouse wheel)
[Gallery not found]
Brief
Bust for showcasing a high fidelity speech synchronisation solution developed by Speech Graphics
(website: http://www.speech-graphics.com )
Use 3d scan data (a base head and 10 blendshapes) to remodel a full bust with higher definition
detail, good topology flow, and a total of 68 blendshapes.
Modelling
Each of the 3d scanned blendshapes had to be re-sculpted to match as close as possible to the
original, while the remaining were all sculpted using only photographic references.
Texturing
The final bust has a normal map that was generated from a very high resolution version of the bust,
a diffuse, specular, glossiness, reflection (for the eyeballs), and opacity (eyelashes) map.
Rigging
I have setup a simple rig to allow eye, jaw, neck, and shoulder movements, skinned the model to the
rig, and created some key animation poses.
Note that the model in these images is for real-time use, and the renders only use 3 simple
spotlights and a blinn material (no global illumination or sophisticated skin shaders).