Stylistic Human Motion Synthesis with Low-dimensional Motion Model and Inverse Kinematics

How to obtain stylistic human motions meeting user-defined constraints is still a challenge in the field of computer animation.To solve this problem,a new low-dimensional motion model that can parameterize motion style precisely is presented in this paper.This model is learnt by analysis of independent feature subspaces analysis.Based on this model,an editing algorithm of stylistic human motion is proposed.This algorithm solves inverse kinematics in the low-dimensional space and responses style parameters in the style subspaces.Users can synthesize stylistic human motions with user-defined style parameters and geometric constrains on key-frames.The experimental results demonstrate the effectiveness of this approach which can be used for interactive motion editing and synthesing.