基于语义虚拟环境的虚拟人动画研究与实现

Research & implementation of virtual human animation based on semantic virtual environment

现有虚拟环境建模(如VRML/X3D)主要集中在几何图形的描述,所显示的虚拟场景对于人是可理解的,但对于计算机是不可理解的,因而难以作有意义的自动处理。基于语义Web设计了一个语义虚拟环境(主要包括语义虚拟场景和语义虚拟人),并设计了基于语义虚拟环境的虚拟人动画引擎。语义虚拟环境使得计算机能够理解和自动处理虚拟环境的语义信息,并且使得虚拟环境系统具有可重用性、可扩展性和可集成性。最后,在语义虚拟环境和虚拟人动画引擎的基础上开发了一个办公室场景中智能虚拟人行动推理的系统,取得了很好的效果。

Modeling of the existing virtual environment（such as VRML/X3D） mainly concentrated in the geometric description,the show of virtual scenes is understandable for people,while is not for computer,so it is difficult to make meaningful automatic processing.Designed a semantic virtual environment（mainly includes semantic virtual scene and semantic virtual human） based on semantic Web,and designed the engine of semantic virtual human animation based on semantic virtual environment.Semantic virtual environments allows computer to understand and automate the semantic information of virtual environment,and makes the virtual environment system can be reused,extended and integrated.Finally,based on the semantic virtual environments and virtual human animation engine,developed an action reasoning system of intelligent virtual human in the office scene,and achieved very good results.