arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1604.05792cs.NE
收藏

用于生成复杂虚拟世界的多智能体进化系统

Multi-agent evolutionary systems for the generation of complex virtual worlds

Jan Kruse, Andy M. Connor

摘要:

现代电影、游戏和虚拟现实应用都依赖于令人信服的计算机图形。高度复杂的模型是成功交付许多场景和环境的必要条件。虽然渲染、合成和动画等工作流程已被简化以适应不断增长的需求,但复杂模型的建模仍然是一项艰巨的任务。本文介绍了计算效益的交互式遗传算法(IGA)的计算机图形建模,同时补偿用户疲劳的影响,一个共同的问题与交互式演化计算。智能代理与IGA结合使用,IGA提供了通过从人类设计师所做的选择中学习并相应地指导搜索来减少用户疲劳的影响的潜力。此工作流可加速基本元素的布局和分布,以形成复杂的模型。它通过互动捕捉设计师的意图,并鼓励有趣的发现。

英文摘要:

Modern films, games and virtual reality applications are dependent on convincing computer graphics. Highly complex models are a requirement for the successful delivery of many scenes and environments. While workflows such as rendering, compositing and animation have been streamlined to accommodate increasing demands, modelling complex models is still a laborious task. This paper introduces the computational benefits of an Interactive Genetic Algorithm (IGA) to computer graphics modelling while compensating the effects of user fatigue, a common issue with Interactive Evolutionary Computation. An intelligent agent is used in conjunction with an IGA that offers the potential to reduce the effects of user fatigue by learning from the choices made by the human designer and directing the search accordingly. This workflow accelerates the layout and distribution of basic elements to form complex models. It captures the designer's intent through interaction, and encourages playful discovery.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑