AI 中文总结
MIRAGE-CAD针对四类输入实现可执行CAD程序的多模态生成,在2500个保留查询上取得构建与STEP导出成功率,证实需分别评估可执行性、几何保真度等指标。
AI 中文摘要
从观测到的对象中恢复可执行参数化CAD程序本质上具有歧义性,因为相同的最终几何形状可由不同的构造过程生成。本研究针对四类输入展开:自然语言描述、渲染图像、点云以及STEP/B-Rep几何。MIRAGE-CAD将每类输入映射至共享的构造表示,并通过显式构造计划接口介导程序生成。生成的Python CAD代码由OpenCASCADE内核执行,以构建实体并将其导出为STEP文件。在每种模态的2500个保留查询上,该系统在推理时未使用检索的情况下,实现了55.4%-70.0%的构建成功率和52.3%-66.2%的STEP导出成功率。受控对比实验表明,强大的重建性能不依赖于将构造表示表达为文本:直接基于连续表示的解码器也能实现强重建,而与曝光匹配的基于计划的解码器在各部分几何保真度上未检测到材料损失。这种显式计划提供了可读且可单独测量的中间表示,其与参考构造的一致性对下游执行成功具有指示意义。最后,研究表明可执行性、几何保真度和参数响应性可能存在显著差异,因此应分别进行评估。
英文摘要
Recovering an executable parametric CAD program from an observed object is fundamentally ambiguous, because the same final geometry can result from different construction procedures. We study this problem from four types of input: natural-language descriptions, rendered images, point clouds, and STEP/B-Rep geometry. MIRAGE-CAD maps each input to a shared construction representation and mediates program generation through an explicit construction-plan interface. The resulting Python CAD code is executed by an OpenCASCADE kernel to build the solid and export it as STEP. On 2,500 held-out queries per modality, the system achieves 55.4-70.0% build success and 52.3-66.2% STEP export success without retrieval at inference. Controlled comparisons show that strong reconstruction does not depend on expressing the construction representation as text: a decoder conditioned directly on the continuous representation also reconstructs strongly, while an exposure-matched plan-based decoder shows no detected material loss in per-part geometric fidelity. The explicit plan instead provides a readable and separately measurable intermediate representation whose agreement with the reference construction is informative about downstream execution success. Finally, we show that executable validity, geometric fidelity, and parametric responsiveness can diverge substantially and should therefore be evaluated separately.
Comments64 pages, 5 figures, 20 tables, 7 appendices. Code, evaluation scripts and run reports: https://github.com/Cad-Kernel/MIRAGE-CAD