2D to 3D Space Transformation for Facial Animation Based on Marker Data

Computer facial animation aims to create an animated character expression as natural as possible as well as human facial expressions. Using the data marker catches facial motion capture, will be determined the location of the feature points of 3D face models to follow the motion of the marker points of human faces. To overcome the morphological differences between the face of the source with the character's face, then applied with radial basis retargeting process mapping so that the character's face can still display the natural expression. Using the data marker 2D, Radial Basis Function (RBF) transformation was applied to determine the position of the feature points on the 3D face models. RBF space transformation has good ability in determining the appropriate facial motion marker points on a human face to the character's face. Motion that occurs in 3D face models is scaled according to the relative scale between the source and the target.