GraphBrep: Learning B-Rep in Graph Structure for Efficient CAD Generation
This work addresses efficiency issues in CAD generation for engineering and design applications, representing an incremental improvement by optimizing computational cost through explicit topology representation.
The paper tackles the challenge of modeling joint distribution of misaligned geometry and topology in direct B-Rep generation for CAD workflows, proposing GraphBrep to explicitly represent compact topology, which reduces training and inference times by up to 31.3% and 56.3% while maintaining high-quality generation compared to state-of-the-art methods.
Direct B-Rep generation is increasingly important in CAD workflows, eliminating costly modeling sequence data and supporting complex features. A key challenge is modeling joint distribution of the misaligned geometry and topology. Existing methods tend to implicitly embed topology into the geometric features of edges. Although this integration ensures feature alignment, it also causes edge geometry to carry more redundant structural information compared to the original B-Rep, leading to significantly higher computational cost. To reduce redundancy, we propose GraphBrep, a B-Rep generation model that explicitly represents and learns compact topology. Following the original structure of B-Rep, we construct an undirected weighted graph to represent surface topology. A graph diffusion model is employed to learn topology conditioned on surface features, serving as the basis for determining connectivity between primitive surfaces. The explicit representation ensures a compact data structure, effectively reducing computational cost during both training and inference. Experiments on two large-scale unconditional datasets and one category-conditional dataset demonstrate the proposed method significantly reduces training and inference times (up to 31.3% and 56.3% for given datasets, respectively) while maintaining high-quality CAD generation compared with SOTA.