BRepCLIP: Contrastive Multimodal Pretraining on BRep Primitives for CAD Understanding

2026-06-03

AI Summary

Existing 3D representation learning methods struggle to effectively model the precise geometry and topology inherent in native CAD boundary representations (B-Reps). This work proposes BRepCLIP, a novel framework that achieves, for the first time, multimodal contrastive alignment between B-Reps and both text and images. The approach serializes CAD models into face and edge tokens enriched with geometric types and semantic descriptors, which are then processed by a Transformer encoder to produce structure-aware global embeddings. These embeddings are jointly pretrained with CLIP to enable cross-modal understanding. Evaluated on the ABC, CADParser, and Automate datasets, BRepCLIP surpasses OpenShape by 40.4%, 22.0%, and 23.9% in Top-1 retrieval accuracy, respectively, and improves zero-shot classification performance on FabWave by 15%. Furthermore, the learned representations serve as an effective evaluation metric for CAD generation tasks.

Abstract

Learning representations of CAD models is a largely open problem. While 3D representation learning has flourished around point clouds and meshes, the native format of CAD - boundary representations BReps, which encodes exact parametric surfaces, curves, and their topology, has received little attention as a representation learning substrate. We introduce BRepCLIP, the first framework to align BRep geometry with language and image embeddings through contrastive pretraining. We model each CAD object as a sequence of face and edge tokens with separate discrete vocabularies for surface and curve geometry, augmented with spatial and semantic descriptors that capture surface types (e.g., cylindrical, torus, NURBS) and curve primitives (e.g., line, arc, B-spline). A transformer encoder aggregates these tokens into a global BRep embedding, aligned with CLIP's text and image encoders via a joint contrastive objective. BRepCLIP generates more discriminative and semantically grounded embeddings than existing point-based alternatives, improving Top-1 retrieval over OpenShape by 40.4%, 22.0%, and 23.9% on ABC, CADParser, and Automate, respectively, and improving zero-shot classification on FabWave by 15% in Top-1 score. We further demonstrate its utility as a CAD-aware similarity metric for evaluating text and image-conditioned CAD generation, establishing the importance of structure-aware pretraining for multimodal CAD understanding. Project page is available at https://muhammadusama100.github.io/BrepClip2026/

Problem

• CAD understanding
• Boundary Representation (BRep)
• Multimodal pretraining
• Representation learning
• Contrastive learning

Innovation

• BRep
• contrastive pretraining
• multimodal CAD understanding
• transformer encoder
• geometric primitives