Figure  14:  High-­‐level  tracks  contain  animation  or  pose  blocks.  

To  implement  co-­‐occurrence,  for  each  joint,  keys  are  found  from  relevant  tracks.  The  

rotations  of  each  influencing  track  are  combined  and  the  new  rotation  is  applied  to  the  joint.    

The  block  structure  follows  a  common  practice  in  animation  studios  where  a  master  

animator  creates  a  dictionary  of  poses  (Thomas  &  Johnston,  1981).    By  encapsulating  

minute  geometric  arrangements  in  concise  groups,  a  master  animator  provides  an  efficient  

mechanism  for  others  to  apply  and  combine  poses.    

In  this  representation,  blocks  can  contain  intensity  curves  that  control  the  onset  and  

intensity  of  a  pose,  allowing  for  subtle  variations.  Joint  rotations  are  multiplied  by  curve  

values  to  vary  the  degree  to  which  each  pose  is  expressed.  Figure  15  demonstrates  the  

animation  curves  in  the  case  of  the  question  “Do  you  want  a  book?”    The  green  curve  

represents  the  movement  corresponding  to  the  yes/no  question  syntactic  marker.    In  ASL,  

the  yes/no  question  marker  is  binary,  thus  the  animation  curve  has  a  quick  onset  and  

relatively  quick  falloff.      The  red  curve  represents  the  influence  of  the  affect  “anger”.    

Emotive  states  are  maintained  for  a  longer  period  of  time  and  are  subject  to  more  variation  

in  intensity.  

  30