submitted to the Workshop on Emotion-based Interaction at conference CASA 2011

In this paper, we present an image-based talking head system that is able to synthesize flexible head motion and realistic facial expression accompanying speech, given arbitrary text input and control tags. The goal of facial animation synthesis is to generate lip synchronized and natural animations. The talking head is evaluated objectively and subjectively.
Demos for three items are evaluated (Please first download the sequences and run them locally for better quality):
1. Objective evaluation of lip synchroniztion

   Subjective Verification of Objective Measurement

   No.	Recorded	Synchronized	Non-synchronized
   1	s001	s001_a	s001_b
   2	s002	s002_a	s002_b
   3	s003	s003_a	s003_b
   4	s004	s004_a	s004_b
   5	s005	s005_a	s005_b

2. Subjective evaluation of facial expression synthesis

   Facial Expression Synthesis

   Pair No.	Recorded	Animated
   1	s001	s001_a
   2	s002	s002_a
   3	s003	s003_a
   4	s004	s004_a
   5	s005	s005_a

   Other exmaples of facial expression synthesis (big smiles):
   big smile 1
   big smile 2

3. Subjective evaluation of head motion synthesis

   Head Motion Synthesis

   Pair No.	Repeated	Flexible
   1	s001	s001_a
   2	s002	s002_a
   3	s003	s003_a
   4	s004	s004_a
   5	s005	s005_a




TNT Home | LUH | Search | Administrator Updated 27/02/2009 Webpage design M.Sc. Kang Liu