A Study on Cooperative Motion Generation for Virtual Actor Team

本研究的目的是分析团队呈现中的词汇，然后开发一个环境，使我们能够在万维网（WWW）上产生“团队呈现”动画。我们首先开发了一个名为T4的自主运动系统，根据捕获的数据动画代理。在2001年的Computer Animation上发表。我们还开发了一个TTS（文本到语音）库MLSLib。它使多个代理人能够同时发言。发表在Journal of WSCG上。另一方面，我们也结合一些文献和真实的电视节目，分析了团队陈述中的词汇。在此基础上，设计了一种团队演示动画语言，并在WWW上进行了开发。实验结果表明，该系统能够很好地生成团队演示动画，证明了该方法的有效性。

The purpose of this research project is to analyze vocabularies in team presentation, and then to develop an environment which enables us to produce "team-presentation" animation on the World Wide Web (WWW).We first developed an autonomous locomotion system named T4 to animate agents based on captured data. It was presented at Computer Animation 2001. We also developed a TTS (Text-To-Speech) library named MLSLib. It enables multiple agents to speak simultaneously. It was presented in Journal of WSCG. We then developed a spatial-temporal-constraint-based cooperative motion generator.On the other hand, we also analyzed vocabularies in team presentation based on some literature and real-world TV programs. We then designed a team-presentation animation language, and we developed it on the WWW.Our experiments showed that the system well produces team-presentation animation and proved the validity of our approach.