Publication
Text2CAD: Generating Sequential CAD Models from Beginner-to-Expert Level Text Prompts
Mohammad Sadil Khan; Sankalp Sinha; Talha Uddin Sheikh; Didier Stricker; Sk Aziz Ali; Muhammad Zeshan Afzal
In: The Thirty-Eighth Annual Conference on Neural Information Processing Systems. Neural Information Processing Systems (NeurIPS-2024), December 10-15, Vancouver, British Columbia, Canada, Neural Information Processing Systems, 12/2024.
Abstract
Prototyping complex computer-aided design (CAD) models in modern software can be very time-consuming. This is due to the lack of intelligent systems that can quickly generate simpler intermediate parts. We propose Text2CAD, the first AI framework for generating text-to-parametric CAD models using designer-friendly instructions for all skill levels. Furthermore, we introduce a data annotation pipeline for generating text prompts based on natural language instructions for the DeepCAD dataset using Mistral and LLaVA-NeXT. The dataset contains ∼ 170K models and ∼ 660K text annotations, from abstract CAD descriptions (e.g.,generate two concentric cylinders) to detailed specifications (e.g., draw two circles with center (x, y) and radius r1, r2, and extrude along the normal by d...). Within the Text2CAD framework, we propose an end-to-end transformer-based auto-regressive network to generate parametric CAD models from input texts. We evaluate the performance of our model through a mixture of metrics, including visual quality, parametric precision, and geometrical accuracy. Our proposed framework shows great potential in AI-aided design applications. Project page is
available at https://sadilkhan.github.io/text2cad-project/