Publication | ACM Transactions on Graphics (SIGGRAPH Proceedings) 2021

Fusion 360 Gallery

A Dataset and Environment for Programmatic CAD Construction from Human Design Sequences

This paper presents the Fusion 360 Gallery reconstruction dataset, containing 8,625 human designed CAD sequences suitable for use with machine learning. The dataset enables learning-based approaches using human CAD design sequences for the first time, with the potential to improve customer workflows such as reverse engineering CAD files from 3D scan data.

Download publication

Abstract

Fusion 360 Gallery: A Dataset and Environment for Programmatic CAD Construction from Human Design Sequences

Karl D.D. Willis, Yewen Pu, Jieliang Luo, Hang Chu, Tao Du, Joseph G. Lambourne, Armando Solar-Lezama, Wojciech Matusik

ACM Transactions on Graphics (SIGGRAPH Proceedings) 2021

Parametric computer-aided design (CAD) is a standard paradigm used to design manufactured objects, where a 3D shape is represented as a program supported by the CAD software. Despite the pervasiveness of parametric CAD and a growing interest from the research community, currently there does not exist a dataset of realistic CAD models in a concise programmatic form. In this paper we present the Fusion 360 Gallery, consisting of a simple language with just the sketch and extrude modeling operations, and a dataset of 8,625 human design sequences expressed in this language. We also present an interactive environment called the Fusion 360 Gym, which exposes the sequential construction of a CAD program as a Markov decision process, making it amendable to machine learning approaches. As a use case for our dataset and environment, we define the CAD reconstruction task of recovering a CAD program from a target geometry. We report results of applying state-of-the-art methods of program synthesis with neurally guided search on this task. Dataset and further information available at: https://github.com/AutodeskAILab/Fusion360GalleryDataset

Related Resources

Publication

2023

CLIP-Forge: Towards Zero-Shot Text-to-Shape Generation

Generating shapes using natural language can enable new ways of…

Publication

2007

The Software That Will Take Digital F/X to the Next Level of Awesome

Jos Stam is standing on a pearl-white beach under a cloudless sky…

Publication

2011

Simulation and the Future of Design Tools for Ecological Research

As buildings mediate between a complex external system (weather) and a…

Publication

2015

Digital Campus Innovation Project: Integration of Building Information Modelling with Building Performance Simulation and Building Diagnostics

Building Information Modelling (BIM) has emerged as a powerful…

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