1. Because we animate in Maya and then light and render in LightWave 3D we have several complete versions of each character - in Maya we have untextured low-detail (proxy) characters for animating with and high-detail untextured versions for translating the animation data to LightWave (using a tool called "The Beaver Project").
In LightWave 3D we have (generally) two versions of each character (different outfits are treated as separate characters) - they both have the same amount of physical geometry but we have two different sets of image mapped textures - high-detail and optimized. The optimized textures are used for the majority of shots and we use the high-detail ones for close-ups and publicity stills.
The objects and textures are kept in a fixed directory structure on our network which breaks down something like this:
Objects/Characters/C_SCA_GEN/HD/C_SCA_GEN_Head.LWO
Where C_SCA_GEN means "Character" "Scarlet" "General Use" - the .LWO indicates that it's a
Light
Wave
Object.
Images are kept in their own folders in a similar fashion:
Images/Characters/C_BLU_GEN/OPT/C_BLU_GEN_Face_RGB.PNG
We do have database software that was written specifically for the production but it was implemented
before LightWave was introduced to the graphics pipeline so it's of limited use to the TD's
2. The studio is split into four teams - Scarlet, Green, White and Blue - each team has their own animators, TD/Lighters, compositors, CG Director and Production Co-ordinator. Consequently, we have no one animator who specifically animates Scarlet or Blue but we do use the same actors for the motion capture and try to make the performances (both mo-capped and animated) as consistent as possible.
3. Whilst a database of mo-cap would indeed be a useful resource, I'm pretty sure that most of the data we use in the show is bespoke and created for each episode - I'm not really the guy to ask though - the animators and mo-cap people would know more about this