HiDiGen: Hierarchical Diffusion for B-Rep Generation with Explicit Topological Constraints
This work addresses the problem of generating valid 3D CAD models for engineering and design applications, representing an incremental improvement by refining existing methods with hierarchical and constraint-based approaches.
The paper tackled the challenge of deep generative modeling of valid Boundary Representation (B-rep) structures in CAD systems by proposing HiDiGen, a hierarchical framework that decouples geometry modeling into two stages with explicit topological constraints, resulting in the generation of novel, diverse, and topologically sound CAD models.
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.