BrepGPT: Autoregressive B-rep Generation with Voronoi Half-Patch

TL;DR

BrepGPT introduces a novel autoregressive framework utilizing Voronoi Half-Patch representation and dual VQ-VAEs to efficiently generate complex boundary representations in CAD models, improving accuracy and versatility.

Contribution

The paper presents a single-stage autoregressive method with a new Voronoi Half-Patch representation for unified B-rep generation, reducing complexity and error accumulation.

Findings

Achieves state-of-the-art performance in unconditional B-rep generation

Demonstrates versatility in conditional generation from various inputs

Enables B-rep autocompletion and interpolation

Abstract

Boundary representation (B-rep) is the de facto standard for CAD model representation in modern industrial design. The intricate coupling between geometric and topological elements in B-rep structures has forced existing generative methods to rely on cascaded multi-stage networks, resulting in error accumulation and computational inefficiency. We present BrepGPT, a single-stage autoregressive framework for B-rep generation. Our key innovation lies in the Voronoi Half-Patch (VHP) representation, which decomposes B-reps into unified local units by assigning geometry to nearest half-edges and sampling their next pointers. Unlike hierarchical representations that require multiple distinct encodings for different structural levels, our VHP representation facilitates unifying geometric attributes and topological relations in a single, coherent format. We further leverage dual VQ-VAEs to encode both vertex topology and Voronoi Half-Patches into vertex-based tokens, achieving a more compact sequential encoding. A decoder-only Transformer is then trained to autoregressively predict these tokens, which are subsequently mapped to vertex-based features and decoded into complete B-rep models. Experiments demonstrate that BrepGPT achieves state-of-the-art performance in unconditional B-rep generation. The framework also exhibits versatility in various applications, including conditional generation from category labels, point clouds, text descriptions, and images, as well as B-rep autocompletion and interpolation.