让AI像人一样一步步重建3D模型，每步都检查结果
CADENA: Stepwise CAD Reverse Engineering
- 分步生成CAD操作序列，每步对比当前几何与目标
- 在多个数据集上超越现有方法，最高提升18.7%
- 适合需要高精度逆向建模的工程师和研究者
计算机辅助设计（CAD）是现代工程的核心，但将现有形状转换为可编辑模型仍需大量专家工作。现有AI系统通常一次性输出整个CAD程序，不检查中间几何结构。相比之下，人类工程师会逐个添加特征，并在每一步后检查剩余待建部分。我们提出CADENA（西班牙语意为“链”），一种逐步重建3D网格为参数化CAD程序的模型，每次生成一个操作步骤，并在每步对比当前预测几何与目标几何。我们还针对机械零件逆向工程缺乏评估基准的问题，提出了CADENA-Bench基准，用于衡量不同类别机械零件的性能。实验表明，CADENA在CADENA-Bench、DeepCAD、Fusion 360和MCB数据集上均优于现有方法。代码开源于https://github.com/zhemdi/cadena，模型权重在https://huggingface.co/kulibinai/cadena，CADENA-Bench数据集可在https://huggingface.co/datasets/kulibinai/cadena-bench获取。
原文摘要 · Abstract (English)
Computer-Aided Design (CAD) underpins modern engineering, yet converting existing shapes into editable models still demands substantial expert effort. Most AI systems emit the entire CAD program in a single pass, never inspecting the intermediate geometry. In contrast, human engineers build a part feature by feature, checking after each operation what remains to be modeled. We introduce CADENA (Spanish for "chain"), a model that reconstructs a 3D mesh as a parametric CAD program, growing its sequence of operations one at a time and comparing the target with the currently predicted geometry at every step. We also address the lack of benchmarks for evaluating reverse-engineering methods on mechanical parts, introducing CADENA-Bench, a benchmark that measures performance across categories of mechanical parts. CADENA outperforms prior methods on CADENA-Bench and on the DeepCAD, Fusion 360, and MCB datasets. Code is available at https://github.com/zhemdi/cadena, model weights at https://huggingface.co/kulibinai/cadena, and CADENA-Bench at https://huggingface.co/datasets/kulibinai/cadena-bench.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。