The Development of an Autonomously Interacting Avatar within a Virtual Reality Environment for Researching the Anticipation Ability in Martial Arts

虚拟现实环境中自主交互化身的开发，用于研究武术中的预期能力

• 批准号: 252070407
• 负责人: Professor Dr. Guido Brunnett
• 金额: --
• 依托单位: Bereich Sportwissenschaft (SPW)
• 依托单位国家: 德国
• 项目类别: Research Grants
• 财政年份: 2014
• 资助国家: 德国
• 起止时间: 2013-12-31 至 2016-12-31
• 项目状态: 已结题
• 来源: https://gepris.dfg.de/gepris/projekt/252070407?language=en
• 关键词: Development; Autonomously; Interacting; Avatar; within

One goal of sports scientific research in the field of anticipation is to create realistic environments. It has been shown that virtual environments that are very similar to real settings evoke realistic reactions in athletes. Previous studies have confirmed that Virtual Reality (VR) can be successfully applied for this purpose. Therein, Virtual Reality has many advantages over the presentation by two dimensional projection screens and computer displays. Amongst the first and foremost advantages of VR is the third dimension that generates a feeling of spatial perception within the users. Further benefits of VR are the high standard of ecological validity, standardization, controllability as well as the possibility to manipulate selected parameters (e.g. temporal and spatial occlusion). However, as far as we know, there is still a big disadvantage in regards to the created virtual opponents. Thus, with these virtual systems it is neither possible to respond to the movements or actions of a real athlete, nor are they interactive. In cooperation with the department of computer science of the Chemnitz University of Technology an interactive virtual opponent shall be developed to overcome these deficits. Based on a rule system, a novel avatar is generated which inherits the necessary short reaction times and which can respond to the real athlete. In addition, the avatar will have a most realistic appearance. From the computer graphics point of view there is still a deficit in the research of the availability of a workflow from the capturing of movement data up to an interactive animation of the virtual opponents skeleton and a realistic presentation of its mesh (the digital humans outward appearance). Moreover, it shall be evaluated in the present project whether a hybrid method of computation (computations done on the graphical processing unit (GPU) and the central processing unit (CPU) simultaneously) can be used to realize our goals.Beyond that, the interactive opponent has to be evaluated in regards to its versatility. More specifically, the evaluation will contain a comparison of reactions of a) an opponent in real life, b) the not interactive opponent and c) the here generated interactive opponent.Finally, the results acquired in the course of the present project will be evaluated concerning the question whether the concept of an autonomously interacting avatar can be successfully adapted to similar types of martial arts.

体育科学研究在预期领域的目标之一是创造现实的环境。已经表明，与真实的设置非常相似的虚拟环境引起运动员的现实反应。以前的研究已经证实，虚拟现实（VR）可以成功地应用于这一目的。其中，虚拟现实具有许多优于二维投影屏幕和计算机显示器的优点。VR的第一个也是最重要的优势之一是第三维，它可以在用户内部产生空间感知的感觉。VR的其他好处是高标准的生态有效性，标准化，可控性以及操纵选定参数（例如时间和空间遮挡）的可能性。然而，据我们所知，在创建虚拟对手方面仍然存在很大的劣势。因此，利用这些虚拟系统，既不可能对真实的运动员的运动或动作做出响应，也不可能是交互式的。在与切姆尼茨理工大学计算机科学系的合作中，将开发一个交互式虚拟对手来克服这些缺陷。基于规则系统，生成新颖的化身，其继承必要的短反应时间并且可以响应于真实的运动员。此外，化身将有一个最现实的外观。从计算机图形学的角度来看，仍然有一个赤字的工作流程的可用性的研究，从捕捉的运动数据的虚拟对手的骨架和一个现实的呈现其网格（数字人的外观）的交互式动画。此外，本项目还需要评估混合计算方法（同时在图形处理器（GPU）和中央处理器（CPU）上进行计算）是否可以用来实现我们的目标。除此之外，还需要评估交互对手的通用性。更具体地，评估将包含a）真实的生活中的对手、B）非交互式对手和c）这里生成的交互式对手的反应的比较。最后，将评估在本项目的过程中获得的结果，该结果涉及自主交互式化身的概念是否可以成功地适用于类似类型的武术的问题。