Unsupervised Image to Sequence Translation with Canvas-Drawer Networks


Unsupervised Image to Sequence Translation with Canvas-Drawer Networks

Kevin Frans, Chin-Yi Cheng

Encoding images as a series of high-level constructs, such as brush strokes or discrete shapes, can often be key to both human and machine understanding. In many cases, however, data is only available in pixel form. We present a method for generating images directly in a high-level domain (e.g. brush strokes), without the need for real pairwise data. Specifically, we train a ”canvas” network to imitate the mapping of high-level constructs to pixels, followed by a high-level ”drawing” network which is optimized through this mapping towards solving a desired image recreation or translation task. We successfully discover sequential vector representations of symbols, large sketches, and 3D objects, utilizing only pixel data. We display applications of our method in image segmentation, and present several ablation studies comparing various configurations.

Download publication

Associated Researchers

Chin-Yi Cheng

Autodesk Research

Kevin Frans

Massachusetts Institute of Technology

View all researchers

Related Resources



Blocks-to-CAD: A Cross-Application Bridge from Minecraft to 3D Modeling

Learning a new software application can be a challenge, requiring the…



Dynamic Opacity Optimization for Scatter Plots

Scatterplots are an effective and commonly used technique to show the…



A Survey of Software Learnability: Metrics, Methodologies and Guidelines

It is well-accepted that learnability is an important aspect of…



WeBuild: Automatically Distributing Assembly Tasks Among Collocated Workers to Improve Coordination

Physical construction and assembly tasks are often carried out by…

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