Automatic Synthesis of Sequences of Human Movements by Linear Combination of Learned Example Patterns
Automatic Synthesis of Sequences of Human Movements by Linear Combination of Learned Example Patterns
复制标题
通过学习示例模式的线性组合自动合成人体运动序列
DOI：
10.1007/3-540-36181-2_54
复制
发表时间：
2002
期刊：
影响因子：
--
通讯作者：
H. Bülthoff
中科院分区：
文献类型：
--
作者：
M. Giese;B. Knappmeyer;H. Bülthoff
我们提出了一种方法，从学习的例子模式的序列逼真的人类运动的合成。我们将此技术应用于动态面部表情的合成。面部运动序列被分解为单个运动元素，这些运动元素通过学习示例的线性组合来建模。线性组合的权重定义了一个抽象的图案空间，该图案空间允许对各个运动元素的风格进行简单的修改和参数化。这些元素的定义方式适合于从具有不同风格的运动元素简单自动重新合成较长序列。我们证明了这种技术的效率为动画的3D头部模型，并讨论它如何可以用来生成时空夸张序列的面部表情的心理物理实验漫画效果。
We present a method for the synthesis of sequences of realistically looking human movements from learned example patterns. We apply this technique for the synthesis of dynamic facial expressions. Sequences of facial movements are decomposed into individual movement elements which are modeled by linear combinations of learned examples. The weights of the linear combinations define an abstract pattern space that permits a simple modification and parameterization of the style of the individual movement elements. The elements are defined in a way that is suitable for a simple automatic resynthesis of longer sequences from movement elements with different styles. We demonstrate the efficiency of this technique for the animation of a 3D head model and discuss how it can be used to generate spatio-temporally exaggerated sequences of facial expressions for psychophysical experiments on caricature effects.