Procedura：具有程序化控制的智能体3D建模

摘要

原生3D生成器现已能从单张图像恢复令人印象深刻的网格几何体。然而，密集网格在机加工物体本应锐利的边缘处仍显柔和，其不具备部件分解结构，也未暴露任何用户可编辑的参数。为解决这一问题，我们探索了“以代码表述三维形状”这一范式，利用并扩展大语言模型（LLM）在三维建模中的编码能力。我们提出了Procedura，一种新颖的三维建模智能体框架，它将物体编写为程序化装配体——一种参数化程序，其中的命名部件通过类型化、机器可检查的配合关系连接。根据文本提示，智能体将物体规划为装配图，并逐部件编写程序，每个部件的放置均基于配合坐标系求解而非猜测，且仅当编译、配合及连接性检查全部通过时才接纳该部件。随后，一个解耦的视觉评判器逐项诊断并修复装配体。此外，同一装配图还承载逐部件材质及经模拟器验证的关节运动。我们在P3D-Bench上依据其装配评判标准进行评估，并采用相同评判标准在自建硬表面基准MechBench-36上进行评测。在两项评测中，Procedura在评判质量上均优于最先进的原生3D生成器及所有此前基于代码的三维智能体，在所评估的方法中产生最锐利的边缘，并且是唯一输出可编辑、具有部件结构的程序的方法。

English

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.