AI 中文总结
针对原生3D生成器缺乏部件结构与可编辑性的问题，提出Procedura框架，以LLM编码能力构建参数化程序集，在P3D-Bench和MechBench-36上表现优于现有方法，生成可编辑的部件结构化3D模型。
AI 中文摘要
原生3D生成器现在能从单张图像恢复出令人印象深刻的网格几何，但密集网格在加工对象本应锐利的地方仍会呈现柔和状态，不包含部件分解，也未提供用户可编辑的参数。为解决这一问题，我们探索将3D形状视为代码的范式，利用并扩展大型语言模型（LLM）的编码能力用于3D建模。我们推出Procedura，一种新型3D建模智能体框架，它将对象编写为程序集，即一种参数化程序，其命名部件通过类型化、机器可检查的配合（mates）连接。智能体从文本提示出发，将对象规划为装配图，并逐部分编写程序，通过配合框架求解每个部件的位置而非猜测，仅在编译、配合和连通性检查通过后才接纳该部件。一个解耦的视觉评论器随后针对每个诊断出的缺陷优化装配。此外，同一图还包含各部件的材质和经模拟器验证的关节结构。我们在P3D-Bench上通过其装配评判器进行评估，并在我们的硬表面基准MechBench-36上使用同一评判器评估。在这两个基准上，Procedura在评判质量上优于最先进的原生3D生成器和所有先前的3D代码智能体，生成了所有评估方法中最锐利的边缘，且是唯一输出可编辑、部件结构化程序的方法。
英文摘要
Native 3D generators now recover impressive mesh geometry from a single image. However, a dense mesh stays soft where a machined object should be sharp, it carries no part decomposition, and it exposes no parameter a user could edit. To address this, we explore the paradigm of 3D shape as code, leveraging and scaling the coding ability of an LLM for 3D modeling. We introduce Procedura, a novel 3D modeling agent framework that writes an object as a procedural assembly, a parametric program whose named parts are joined by typed, machine-checkable mates. From a text prompt, the agent plans the object as an assembly graph and writes the program part by part, solving each placement from the mated frames rather than guessing it, and admitting a part only once compile, mate, and connectivity checks pass. A decoupled vision critic then refines the assembly one diagnosed fix at a time. Moreover, the same graph carries per-part materials and a simulator-validated articulation. We evaluate on P3D-Bench under its assembly judge, and with the same judge on MechBench-36, our hard-surface benchmark. On both, Procedura outperforms state-of-the-art native 3D generators and every prior 3D-code agent on judged quality, produces the sharpest edges of any method we evaluate, and is the only one whose output is an editable, part-structured program.
CommentsProject page: https://spatiaos.github.io/projects/procedura/
