Inverse 3D Modeling

逆向 3D 建模

• 批准号: 1029662
• 负责人: Carlo Sequin
• 金额: $ 37.4万
• 依托单位: University of California-Berkeley
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2010
• 资助国家: 美国
• 起止时间: 2010-09-01 至 2013-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1029662&HistoricalAwards=false
• 关键词: Inverse; 3D; Modeling

The research objective of this award is to find new, generally applicable ways to facilitate the process of re-creating and modifying complex geometrical shapes that exist only in an un-structured, low-level format -- without a high-level description of its design process or of the techniques by which the original shape had been designed. The purpose is to speed up the redesign of engineering parts, consumer goods, or artistic items -- a process that often starts from a suitable precursor objects and then applies specific modifications to meet the new design objectives. An integrated system of extraction modules will be developed that turns a hierarchically flat geometry description, such as a boundary representation or a cloud of points, into a structured, higher-level representation, comprising a reasonable set of adjustable parameters. This higher-level procedural description then serves as a good starting point to introduce design modifications and enhancements. Given that the possible decompositions of a given geometrical shape are neither obvious nor unique, and the generic underlying forms may be obscured by added-on surface details or by noise from a scanning process, and that the best procedural description for a given object depends strongly on the intended design modifications,the designer will be included in this Inverse 3D Modeling process. If successful, this research will result in a new approach to shape design for many diverse application domains. The use of appropriately parameterized descriptions as a starting point will make re-design efforts more efficient and thus reduce time to market for the revised product. The availability of such Inverse 3D Modeling tools for re-design purposes will enhance the competitiveness of any industry branch that employs them. Moreover, understanding the fundamental operations used to generate a model, and exposing those operations in a flexible, generative way, also enhances the creative process itself. A wider impact of the envisioned modular Inverse 3D Modeling system lies in education. This approach will be used in courses on geometric design and modeling, thereby exposing students quickly to several different ways of understanding a given shape from the point of view of generating it procedurally. This broadens the experience of the students and gives them a rich repertoire of approaches to draw from, when they are faced with novel modeling tasks. Outreach activities will also focus on high-school students, who will make use of this fun computer tool to increase interest in science and engineering, and, hopefully, influence the students' career choices.

该奖项的研究目标是找到新的，普遍适用的方法，以促进重新创建和修改复杂的几何形状的过程中，只存在于一个非结构化的，低层次的格式-没有一个高层次的描述其设计过程或原始形状的设计技术。其目的是加快工程部件、消费品或艺术品的重新设计--这一过程通常从合适的前体对象开始，然后进行特定的修改以满足新的设计目标。将开发提取模块的集成系统，该系统将分层平坦的几何描述（例如边界表示或点云）转变为结构化的更高级别的表示，包括一组合理的可调整参数。然后，这种更高级别的过程描述可以作为引入设计修改和增强的良好起点。鉴于给定几何形状的可能分解既不明显也不独特，并且通用的潜在形式可能会被附加的表面细节或扫描过程中的噪声所掩盖，并且给定对象的最佳程序描述在很大程度上取决于预期的设计修改，设计师将被包括在此逆向3D建模过程中。如果成功的话，这项研究将为许多不同的应用领域提供一种新的形状设计方法。使用适当的参数化描述作为起点将使重新设计工作更有效，从而缩短修订产品的上市时间。这种用于重新设计目的的逆向3D建模工具的可用性将增强使用它们的任何行业分支的竞争力。此外，理解用于生成模型的基本操作，并以灵活的生成方式公开这些操作，也可以增强创造性过程本身。设想的模块化逆向3D建模系统的更广泛的影响在于教育。这种方法将用于几何设计和建模课程，从而使学生快速了解从程序生成的角度理解给定形状的几种不同方法。这拓宽了学生的经验，并为他们提供了丰富的方法库，当他们面临新的建模任务时。外展活动也将侧重于高中生，他们将利用这一有趣的计算机工具来提高对科学和工程的兴趣，并希望影响学生的职业选择。