Real-time 3D facial expression control based on performance

Performance-driven facial expression animation is one of the researching highlight in computer animation. In this paper, a method is proposed to control 3D facial expression animations in real time. The skeleton based on skinning was adopted for an efficient control of the deformation of the face model. Kinect was used as a capture device to get the depth and color images of the performers that help to obtain the animation units in candide-3. The animation unites smoothed by moving average filter were used to control the translation and rotation of skeletons, which produced a good result of the facial animations.

[1]  Yu Zhang.  Muscle-driven modeling of wrinkles for 3D facial expressions , 2008, 2008 IEEE International Conference on Multimedia and Expo.

[2]  Ming Zhang,et al.  Real-time performance-driven facial animation with 3ds Max and Kinect , 2013, 2013 3rd International Conference on Consumer Electronics, Communications and Networks.

[3]  Hao Li,et al.  Realtime performance-based facial animation , 2011, ACM Trans. Graph..

[4]  Jörgen Ahlberg,et al.  CANDIDE-3 - An Updated Parameterised Face , 2001 .

[5]  Saïda Bouakaz,et al.  Easy acquisition and real‐time animation of facial wrinkles , 2011, Comput. Animat. Virtual Worlds.

[6]  Xiaopeng Wei,et al.  Facial Animation Based on Feature Points , 2013 .

[7]  F. I. Parke June,et al.  Computer Generated Animation of Faces , 1972 .

[8]  Frederic I. Parke,et al.  A parametric model for human faces. , 1974 .

[9]  Keith Waters,et al.  A muscle model for animation three-dimensional facial expression , 1987, SIGGRAPH.

[10]  Ziling Su,et al.  3D Face Reconstruction Based on Improved CANDIDE-3 Model , 2012, 2012 Fourth International Conference on Digital Home.

[11]  Xu Xu Chen,et al.  Margin Refine of Candide-3 Model for Facial Deformation , 2014 .

[12]  Kun Zhou,et al.  3D shape regression for real-time facial animation , 2013, ACM Trans. Graph..