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

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Li, Pu, Zhang, Wenhao, Quan, Weize, Zhang, Biao, Wonka, Peter, Yan, Dong-Ming
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914172594290688
author Li, Pu
Zhang, Wenhao
Quan, Weize
Zhang, Biao
Wonka, Peter
Yan, Dong-Ming
author_facet Li, Pu
Zhang, Wenhao
Quan, Weize
Zhang, Biao
Wonka, Peter
Yan, Dong-Ming
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2511_22171
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle BrepGPT: Autoregressive B-rep Generation with Voronoi Half-Patch
Li, Pu
Zhang, Wenhao
Quan, Weize
Zhang, Biao
Wonka, Peter
Yan, Dong-Ming
Computer Vision and Pattern Recognition
Graphics
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.
title BrepGPT: Autoregressive B-rep Generation with Voronoi Half-Patch
topic Computer Vision and Pattern Recognition
Graphics
url https://arxiv.org/abs/2511.22171