Publication | International Conference on Machine Learning 2022
SkexGen
Autoregressive Generation of CAD Construction Sequences with Disentangled Codebooks
This Autodesk Research paper describes a new approach to generation of solid CAD models that enhances user control and enables efficient exploration of the design space.
Download publicationAbstract
SkexGen: Autoregressive Generation of CAD Construction Sequences with Disentangled Codebooks
Xiang Xu, Karl D.D. Willis, Joseph G. Lambourne, Chin-Yi Cheng, Pradeep Kumar Jayaraman, Yasutaka Furukawa
International Conference on Machine Learning 2022
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.
Associated Researchers
Chin-Yi Cheng
Autodesk Research
Yasutaka Furukawa
Simon Fraser University
Related Resources
2024
GraphSeam: Supervised Graph Learning Framework for Semantic UV MappingProposing a fully automated UV mapping framework that enables users to…
2023
SolidGen: An Autoregressive Model for Direct B-rep SynthesisA generative model that can synthesize 3D CAD models in the boundary…
2022
COIL: Constrained Optimization in Workshop on Learned Latent SpaceConstrained optimization problems can be difficult because their…
2022
JoinABLe: Learning Bottom-up Assembly of Parametric CAD JointsPhysical products are often complex assemblies combining a multitude…
Get in touch
Something pique your interest? Get in touch if you’d like to learn more about Autodesk Research, our projects, people, and potential collaboration opportunities.
Contact us