Maya offers excellent functional flexibility but often suffers from sub-optimal performance scalability as a counterpart. This becomes problematic in large productions, where tens of artists work together on a single scene. One scene file can easily contain tens of thousands of nodes. Here we share our experience of improving the performance of highly complex facial rigs in Maya.
Jaewoo Seo is a computer graphics researcher. His research interests include motion capture and animation techniques of faces and bodies. He has presented a number of research papers in international journals and conferences such as Siggraph or Siggraph Asia. At Weta Digital, he has been working on several R&D projects in facial animation and body motion capture for last two years. Before coming to New Zealand, he was a researcher at OLM Digital, Inc., one of the leading animation and visual effects companies in Japan. He received his PhD (2012) and MS (2007) degrees in Culture Technology from KAIST, Korea and Bachelor degrees in Digital Media and Computer Science (both in 2005) from Ajou University, Korea.