SketchDNN: Joint Continuous-Discrete Diffusion for CAD Sketch Generation
This work addresses CAD sketch generation for design automation, establishing a new state-of-the-art, though it is incremental as it builds on diffusion models for a specific domain.
The paper tackled the problem of generating CAD sketches by proposing SketchDNN, a generative model that jointly models continuous parameters and discrete class labels using a unified diffusion process, resulting in improved generation quality with FID reduced from 16.04 to 7.80 and NLL from 84.8 to 81.33.
We present SketchDNN, a generative model for synthesizing CAD sketches that jointly models both continuous parameters and discrete class labels through a unified continuous-discrete diffusion process. Our core innovation is Gaussian-Softmax diffusion, where logits perturbed with Gaussian noise are projected onto the probability simplex via a softmax transformation, facilitating blended class labels for discrete variables. This formulation addresses 2 key challenges, namely, the heterogeneity of primitive parameterizations and the permutation invariance of primitives in CAD sketches. Our approach significantly improves generation quality, reducing Fréchet Inception Distance (FID) from 16.04 to 7.80 and negative log-likelihood (NLL) from 84.8 to 81.33, establishing a new state-of-the-art in CAD sketch generation on the SketchGraphs dataset.