CADKnitter: Compositional CAD Generation from Text and Geometry Guidance

TL;DR

CADKnitter is a novel framework that generates multi-part CAD models from text prompts and geometric constraints, significantly advancing automated and editable CAD design processes.

Contribution

It introduces a compositional CAD generation method with a geometry-guided diffusion strategy and curates a large dataset, KnitCAD, for training and evaluation.

Findings

Outperforms state-of-the-art baselines in CAD generation tasks.

Successfully generates CAD models adhering to both semantic and geometric constraints.

Demonstrates effectiveness on a large dataset of over 310,000 samples.

Abstract

Crafting computer-aided design (CAD) models has long been a painstaking and time-intensive task, demanding both precision and expertise from designers. With the emergence of 3D generation, this task has undergone a transformative impact, shifting not only from visual fidelity to functional utility but also enabling editable CAD designs. Prior works have achieved early success in single-part CAD generation, which is not well-suited for real-world applications, as multiple parts need to be assembled under semantic and geometric constraints. In this paper, we propose CADKnitter, a compositional CAD generation framework with a geometry-guided diffusion sampling strategy. CADKnitter is able to generate a complementary CAD part that follows both the geometric constraints of the given CAD model and the semantic constraints of the desired design text prompt. We also curate a dataset, so-called KnitCAD, containing over 310,000 samples of CAD models, along with textual prompts and assembly metadata that provide semantic and geometric constraints. Intensive experiments demonstrate that our proposed method outperforms other state-of-the-art baselines by a clear margin.