HiDiGen: Hierarchical Diffusion for B-Rep Generation with Explicit Topological Constraints

Abstract

Boundary representation (B-rep) is the standard 3D modeling format in CAD systems, encoding both geometric primitives and topological connectivity. Despite its prevalence, deep generative modeling of valid B-rep structures remains challenging due to the intricate interplay between discrete topology and continuous geometry. In this paper, we propose HiDiGen, a hierarchical generation framework that decouples geometry modeling into two stages, each guided by explicitly modeled topological constraints. Specifically, our approach first establishes face-edge incidence relations to define a coherent topological scaffold, upon which face proxies and initial edge curves are generated. Subsequently, multiple Transformer-based diffusion modules are employed to refine the geometry by generating precise face surfaces and vertex positions, with edge-vertex adjacencies dynamically established and enforced to preserve structural consistency. This progressive geometry hierarchy enables the generation of more novel and diverse shapes, while two-stage topological modeling ensures high validity. Experimental results show that HiDiGen achieves strong performance, generating novel, diverse, and topologically sound CAD models.

Figures (9)

• Figure 1: Comparison of learning paradigms for B-rep generation: (a) implicit topology encoding, (b) decoupled topology-geometry generation, and (c) our hierarchical generation framework.
• Figure 2: HiDiGen pipeline overview. Our framework generates B-rep models in two hierarchical levels: Level 1 produces face-edge topology ($\mathbf{EF}_i$) and coarse geometry (face bounding boxes $\mathbf{B}_i$, edge curves $\mathbf{E}_i$); Level 2 refines edge-vertex topology ($\mathbf{EV}_i$) and detailed geometry (face surfaces $\mathbf{F}_i$, vertex positions $\mathbf{V}_i$), with each stage conditioned on prior outputs.
• Figure 3: Illustration of the contextual conditioned embedding for face geometry generation. We gather the corresponding face bounding boxes and edge geometries, and further derive topological connectivity relationships to form the conditioning embedding.
• Figure 4: Qualitative comparison of unconditioned B-rep models generated by our method and prior approaches, including DeepCAD wu2021deepcad, BrepGen xu2024brepgen, and DTGBrepGen li2025dtgbrepgen, on the DeepCAD dataset. Our method generates more versatile and consistently watertight solid B-reps, demonstrating geometric fidelity and topological correctness.
• Figure 5: HiDiGen failure cases illustrating error propagation between topology and geometry: (a) erroneous topology yields incorrect geometry; (b) inaccurate geometry leads to spurious topological connections.