Performance-Driven Two-Dimensional Vector Expression Animation

To produce vivid expression animation quickly and conveniently,we present a new algorithm to retarget the motion capture data onto the cartoon face.The algorithm allows users mark the cartoon face with the same topology as the expression data,and compares the first frame of the expression data with the marked control points.The resulting ratio is used to retarget the rest of the expression data onto the cartoon face,yielding the right expression data.The experimental results show that our approach is practical to generate vivid expression animation.