DeepCAD is a CAD dataset consisting of 179,133 models and their CAD construction sequences. It can be used to train generative models of 3D shapes.

The Fusion 360 Gallery Dataset contains rich 2D and 3D geometry data derived from parametric CAD models. The dataset is produced from designs submitted by users of the CAD package Autodesk Fusion 360 to the Autodesk Online Gallery. The dataset provides valuable data for learning how people design, including sequential CAD design data, designs segmented by modelling operation, and design hierarchy and connectivity data. Source: https://github.com/AutodeskAILab/Fusion360GalleryDataset Image Source: https://github.com/AutodeskAILab/Fusion360GalleryDataset

The CC3D dataset [1] of 3D CAD models was collected from a free online service for sharing CAD designs [2]. In total, the collected dataset contains 50k+ models, unrestricted to any category, with varying complexity from simple to highly detailed designs. These CAD models are converted to meshes, and each mesh was virtually scanned using a proprietary 3D scanning pipeline developed by Artec3D [3]. The typical size of the resulting scans is in the order of 100K points and faces, while the meshes converted from CAD models are usually more than an order of magnitude lighter. The availability of CAD-3D scan pairings, the high-resolution of meshes, and the variability of the models make the CC3D dataset stand out among other alternatives. [1] Pvdeconv: Point-voxel deconvolution for autoencoding cad construction in 3d, Cherenkova, Kseniya, Djamila Aouada, and Gleb Gusev, 2020 IEEE International Conference on Image Processing (ICIP). [2] “3dcontentcentral,” https://www.3dcontentcentral.com. [3] “Artec3d,” https://www.artec3d.com/.

Prototyping complex computer-aided design (CAD) models in modern softwares 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 $\sim170$K models and $\sim660$K text annotations, from abstract CAD descriptions (e.g., \textit{generate two concentric cylinders}) to detailed specifications (e.g., \textit{draw two circles with center} $(x,y)$ \textit{and radius} $r_{1}$, $r_{2}$, \textit{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~\href{https://sadilkhan.github.io/text2cad-project/}{https://sadilkhan.github.io/text2cad-project/}.

📚 BlendNet The dataset contains $12k$ samples. To balance cost savings with data quality and scale, we manually annotated $2k$ samples and used GPT-4o to annotate the remaining $10k$ samples. For more details, please visit our GitHub repository

📚 CADBench CADBench is a comprehensive benchmark to evaluate the ability of LLMs to generate CAD scripts. It contains 500 simulated data samples and 200 data samples collected from online forums. For more details, please visit our GitHub repository