| 摘要 |
Natural inter-viseme animation of 3D head model driven by speech recognition is calculated by applying limitations to the velocity and/or acceleration of a normalized parameter vector, each element of which may be mapped to animation node outputs of a 3D model based on mesh blending and weighted by a mix of key frames.
|
