CAREER: Understanding the Complexities of Animated Content

职业：了解动画内容的复杂性

• 批准号: 0237706
• 负责人: Ronald Metoyer
• 金额: $ 49.37万
• 依托单位: Oregon State University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2003
• 资助国家: 美国
• 起止时间: 2003-03-15 至 2010-02-28
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0237706&HistoricalAwards=false
• 关键词: CAREER; Understanding; Complexities; Animated; Content

CAREER: Understanding the Complexities of Animated ContentAbstractAs immersive graphics environments become more viable as useful training and education tools, we must be able to understand animated content well enough to provide naive content creators with the ability to create compelling synthetic scenes. This understanding requires that we answer the following questions. How do we create characters that perform natural motions based on high-level direction? How do we direct synthetic characters at a high level? How do we structure interactive spaces for spatial and temporal interactions between humans and synthetic characters?To answer these questions, we will conduct research in the following areas: motion capture re-sequencing, character interfaces and character programming. We will restrict ourselves to a particular domain and build an interactive space in order to study the above research questions. We have chosen quarterback training for the sport of American football as the specific problem domain for the following reasons. Quarterback training requires compelling 3D content in order to produce convincing training situations. Quarterback training is a physical task and involves spatial and temporal interaction between the coaches and players (where players include the real quarterback and the synthetic characters). Game preparation requires time-critical training content creation. Quarterback training requires use by domain experts (coaches and players) who may be naive computer users. We have access to domain experts in quarterback training through the Oregon State University football program. Our research will result in algorithms for generating natural motion from motion capture data based on high level input, guidelines for classifying character interaction techniques, and methods for directing characters within the interactive space. We plan to evaluate our research through various quantitative measures and usability studies to determine the quality of content and the ability of naive content creators to create and interact with the content within our interactive space framework.

职业：了解动画内容的复杂性摘要随着沉浸式图形环境作为有用的培训和教育工具变得更加可行，我们必须能够充分理解动画内容，以便为天真的内容创建者提供创建引人注目的合成场景的能力。这种理解要求我们回答以下问题。我们如何创建基于高级方向执行自然动作的角色？我们如何在高水平上指导合成角色？我们如何构建人类与合成角色之间时空交互的交互空间？为了回答这些问题，我们将在以下领域进行研究：动作捕捉重排序、角色界面和角色编程。我们将把自己限制在一个特定的领域，构建一个互动的空间来研究上述的研究问题。我们选择美式橄榄球运动的四分卫训练作为具体问题领域的原因如下。四分卫训练需要引人注目的 3D 内容才能产生令人信服的训练场景。四分卫训练是一项身体任务，涉及教练和球员之间的空间和时间互动（其中球员包括真正的四分卫和合成角色）。游戏准备需要时间紧迫的培训内容创建。四分卫训练需要由领域专家（教练和球员）使用，他们可能是初级计算机用户。我们可以通过俄勒冈州立大学橄榄球项目接触到四分卫训练领域的专家。我们的研究将产生基于高级输入的动作捕捉数据生成自然动作的算法、角色交互技术分类指南以及在交互空间内引导角色的方法。我们计划通过各种定量措施和可用性研究来评估我们的研究，以确定内容的质量以及内容创作者在我们的互动空间框架内创建内容并与之互动的能力。