Intelligent Expression Blending in Performance Driven Facial Animation 12+

📘 Intelligent Expression Blending in PDFA Performance Driven Facial Animation (PDFA) provides a cost-effective and intuitive means of creating facial avatars. In Vision based PDFA, facial actions of an input facial video are re-targeted to desired facial model. However, PDFA lacks facilities for intelligent editing of the re- targeted video. This book discusses how to incorporate flexibility to PDFA so that resulting system can intelligently edit the input video on basis of contextual information. It advocates the use of Fuzzy Theory to automatically generate the facial actions that must be blended with the input video's facial actions to truly reflect the current context. These changes can then be translated into facial action parameters for synthesizing context- based facial expressions of the avatar.