ShapeMOD: macro operation discovery for 3D shape programs
ShapeMOD: macro operation discovery for 3D shape programs
复制标题
ShapeMOD：3D 形状程序的宏操作发现
DOI：
10.1145/3450626.3459821
复制
发表时间：
2021
影响因子：
6.2
通讯作者：
Ritchie, Daniel
中科院分区：
文献类型：
--
作者：
Jones, R. Kenny;Charatan, David;Guerrero, Paul;Mitra, Niloy J.;Ritchie, Daniel
创建详细且易于控制的 3D 形状的一种流行方法是通过程序建模，即使用程序生成几何图形。此类程序由一系列指令及其相关参数值组成。为了充分实现这种表示的好处，形状程序应该是紧凑的，并且仅公开允许对输出几何体进行有意义的操作的自由度。实现这一目标的一种方法是设计更高级别的宏操作器，这些操作器在执行时会扩展为来自基本形状建模语言的一系列命令。然而，手动编写此类宏，就像形状程序本身一样，很困难，并且很大程度上仅限于领域专家。在本文中，我们提出了 ShapeMOD，一种用于自动发现在 3D 形状程序的大型数据集中有用的宏的算法。 ShapeMOD 运行在以命令式、基于语句的语言表达的形状程序上。它旨在发现宏，通过最大限度地减少表示输入形状集合所需的函数调用和自由参数的数量，使程序更加紧凑。我们在以 3D 形状结构的领域特定语言表达的多个程序集合上运行 ShapeMOD。我们证明它会自动发现一组简洁的宏，这些宏可以抽象出通用的结构和参数模式，这些模式可以泛化到大型形状集合上。我们还证明了 ShapeMOD 发现的宏可以提高下游任务的性能，包括形状生成建模和从点云推断程序。最后，我们进行了一项用户研究，表明 ShapeMOD 发现的宏使交互式形状编辑更加高效。
A popular way to create detailed yet easily controllable 3D shapes is via procedural modeling, i.e. generating geometry using programs. Such programs consist of a series of instructions along with their associated parameter values. To fully realize the benefits of this representation, a shape program should be compact and only expose degrees of freedom that allow for meaningful manipulation of output geometry. One way to achieve this goal is to design higher-levelmacrooperators that, when executed, expand into a series of commands from the base shape modeling language. However, manually authoring such macros, much like shape programs themselves, is difficult and largely restricted to domain experts. In this paper, we present ShapeMOD, an algorithm for automatically discovering macros that are useful across large datasets of 3D shape programs. ShapeMOD operates on shape programs expressed in an imperative, statement-based language. It is designed to discover macros that make programs more compact by minimizing the number of function calls and free parameters required to represent an input shape collection. We run ShapeMOD on multiple collections of programs expressed in a domain-specific language for 3D shape structures. We show that it automatically discovers a concise set of macros that abstract out common structural and parametric patterns that generalize over large shape collections. We also demonstrate that the macros found by ShapeMOD improve performance on downstream tasks including shape generative modeling and inferring programs from point clouds. Finally, we conduct a user study that indicates that ShapeMOD's discovered macros make interactive shape editing more efficient.