三维人体模型的自动驱动方法

Method of drive 3D model to move automatically

传统的由运动数据驱动三维模型需要手工标定骨架、匹配关节点和设定皮肤权重,工作量大而复杂。提出了三维人体模型自动驱动的方法,使用横向切分匹配来自动提取骨架和匹配相应的关节点,然后用改进后的线性混合皮肤变形技术进行模型皮肤权值的自动分配,绑定皮肤,最终实现了三维人体模型的自动驱动。整个提取绑定时间短,需求简单,速度快,最后给出的实验结果验证了算法的有效性。

When we use some traditional methods to drive a 3D model to move based its 3D motion parameters, the workload become great and complex due to identify the skeleton, match the joint and set skin weight manually. A novel method is presented, which can drive a 3D model to move automatically. Firstly, a horizontal section matching method is used to extract the skeleton of the 3D human model and match these corresponding model joints automatically. Secondly, an improved linear blend skinning algorithm is performed to assign the weight of the model skin automatically and blind the skin. Finally, a 3D human model is driven automatically： The entire extraction and match time is short, the demand for user is simple. At end of this paper, several experiments are given to demonstrate the proposed method is efficient.