Better STEP, a format and dataset for boundary representation

Boundary representation (B-rep) generated from computer-aided design (CAD) is widely used in industry, with several large datasets available. However, the data in these datasets is represented in STEP format, requiring a CAD kernel to read and process it. This dramatically limits their scope and usage in large learning pipelines, as it constrains the possibility of deploying them on computing clusters due to the high cost of per-node licenses. This dataset introduces an alternative format based on the open, cross-platform format HDF5 and a corresponding dataset for STEP files. To demonstrate the effectiveness of the format, we converted the Fusion 360 dataset and the ABC dataset. For more information visit https://better-step.github.io从计算机辅助设计（Computer-aided design, CAD）生成的边界表示法（Boundary representation, B-rep）在工业界应用广泛，目前已有多个大型公开数据集面世。然而，此类数据集的数据均采用STEP格式存储，需借助CAD内核完成读取与处理操作。这极大限制了其在大型机器学习流水线中的应用范畴与使用场景，由于单节点许可证成本高昂，将其部署至计算集群的可行性也受到严重约束。本数据集提出了一种基于开源跨平台格式HDF5的替代存储格式，并配套提供了针对STEP文件的转换数据集。为验证该格式的有效性，我们对Fusion 360数据集与ABC数据集完成了格式转换。更多详情可访问 https://better-step.github.io

搜集汇总

背景与挑战

背景概述

该数据集提供了一种基于HDF5的边界表示（B-rep）格式和对应数据集，旨在解决传统STEP格式在CAD数据中因依赖CAD内核而导致的处理限制和许可成本问题。它通过转换Fusion 360和ABC数据集，实现了更紧凑、高效的存储和访问，适用于大规模学习管道，总大小为551.5 GB，仅用于非商业学术研究。