BRepCLIP: Contrastive Multimodal Pretraining on BRep Primitives for CAD Understanding

Hugging Face Daily Papers Papers

Summary

BRepCLIP introduces contrastive multimodal pretraining on boundary representation (BRep) primitives for CAD understanding, aligning BRep geometry with language and image embeddings to achieve state-of-the-art retrieval and zero-shot classification.

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/
Original Article
View Cached Full Text

Cached at: 06/05/26, 10:10 PM

Paper page - BRepCLIP: Contrastive Multimodal Pretraining on BRep Primitives for CAD Understanding

Source: https://huggingface.co/papers/2606.05515

Abstract

BRepCLIP enables multimodal representation learning for CAD models by aligning boundary representation geometry with language and image embeddings through contrastive pretraining, achieving superior retrieval and classification performance compared to point-based methods.

Learning representations ofCAD modelsis a largely open problem. While 3D representation learning has flourished around point clouds and meshes, the native format of CAD -boundary representationsBReps, 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 throughcontrastive pretraining. We model each CAD object as a sequence of face andedge tokenswith separatediscrete vocabulariesfor surface and curve geometry, augmented with spatial andsemantic descriptorsthat capture surface types (e.g., cylindrical, torus, NURBS) and curve primitives (e.g., line, arc, B-spline). Atransformer encoderaggregates these tokens into aglobal BRep embedding, aligned withCLIP’s text and image encoders via a joint contrastive objective. BRepCLIPgenerates more discriminative and semantically grounded embeddings than existing point-based alternatives, improvingTop-1 retrievalover OpenShape by 40.4%, 22.0%, and 23.9% on ABC, CADParser, and Automate, respectively, and improvingzero-shot classificationon 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 formultimodal CAD understanding. Project page is available at https://muhammadusama100.github.io/BrepClip2026/

View arXiv pageView PDFProject pageAdd to collection

Get this paper in your agent:

hf papers read 2606\.05515

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2606.05515 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2606.05515 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2606.05515 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

CLIP: Connecting text and images

OpenAI Blog

CLIP is OpenAI's vision-language model that learns from text-image pairs from the internet, enabling zero-shot visual classification without task-specific training data. It addresses major limitations in traditional computer vision by reducing dependence on expensive labeled datasets and improving real-world generalization.

KODA: Contrastive Representation Comparison and Alignment for Vision-Language Foundation Models

arXiv cs.LG

This paper introduces KODA (Kernel Optimization for Discrepancy Analysis), a kernel-based framework for comparing and aligning vision-language model representations by identifying sample subsets that are clustered differently across models like CLIP, SigLIP, and BLIP. The method uses contrastive embedding clustering and randomized low-dimensional approximations to scale to large datasets while providing interpretable structural differences between representations.

Balancing Multimodal Learning through Label Space Reshaping

arXiv cs.LG

Introduces Balanced Multimodal Label Reshaping (BMLR), a method that addresses modality imbalance in multimodal learning by reshaping the label space to equalize mapping difficulty across modalities, improving performance across various architectures.