SkexGen: Autoregressive Generation of CAD Construction Sequences with Disentangled Codebooks

2022/07/11 by Xiang Xu, Karl D. D. Willis, Xu, Xiang +9 · 17 citations
Engineering · #3D Shape Modeling and Analysis #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Innovations in Concrete and Construction Materials #Machine Learning (cs.LG) #Manufacturing Process and Optimization

paper · pdf · doi:10.48550/arxiv.2207.04632

openalex publication_date 2022/07/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We present SkexGen, a novel autoregressive generative model for computer-aided design (CAD) construction sequences containing sketch-and-extrude modeling operations. Our model utilizes distinct Transformer architectures to encode topological, geometric, and extrusion variations of construction sequences into disentangled codebooks. Autoregressive Transformer decoders generate CAD construction sequences sharing certain properties specified by the codebook vectors. Extensive experiments demonstrate that our disentangled codebook representation generates diverse and high-quality CAD models, enhances user control, and enables efficient exploration of the design space. The code is available at https://samxuxiang.github.io/skexgen.

Cited by

• Wrong Design Intent Is Worse Than None: A Derangement-Control Diagnosis of Header Conditioning in CAD Program Completion
• GLUE: Coordinating Pre-Trained Generative Models for System-Level Design
• CADMorph: Geometry-Driven Parametric CAD Editing via a Plan-Generate-Verify Loop
• CADKnitter: Compositional CAD Generation from Text and Geometry Guidance
• AutoBrep: Autoregressive B-Rep Generation with Unified Topology and Geometry
• BrepGPT: Autoregressive B-rep Generation with Voronoi Half-Patch
• MamTiff-CAD: Multi-Scale Latent Diffusion with Mamba+ for Complex Parametric Sequence
• B-Rep Distance Functions (BR-DF): How to Represent a B-Rep Model by Volumetric Distance Functions?
• Target-Guided Bayesian Flow Networks for Quantitatively Constrained CAD Generation
• GACO-CAD: Geometry-Augmented and Conciseness-Optimized CAD Model Generation from Single Image
• CAD-Tokenizer: Towards Text-based CAD Prototyping via Modality-Specific Tokenization
• IndustryForge-27B: A Domain-Enhanced Multimodal Foundation Model for Industrial CAD
• Generating Human-AI Collaborative Design Sequence for 3D Assets via Differentiable Operation Graph
• GenCAD-3D: CAD Program Generation using Multimodal Latent Space Alignment and Synthetic Dataset Balancing
• A Design Co-Pilot for Task-Tailored Manipulators
• Drawing2CAD: Sequence-to-Sequence Learning for CAD Generation from Vector Drawings
• CADDesigner: Conceptual Design of CAD Models Based on General-Purpose Agent

Related

• CADmium: Fine-Tuning Code Language Models for Text-Driven Sequential CAD Design
• MamTiff-CAD: Multi-Scale Latent Diffusion with Mamba+ for Complex Parametric Sequence
• Fusion 360 Gallery: A Dataset and Environment for Programmatic CAD Construction from Human Design Sequences
• Hierarchical Neural Coding for Controllable CAD Model Generation
• DeepCAD: A Deep Generative Network for Computer-Aided Design Models