Generative Layout Modeling using Constraint Graphs

TL;DR

This paper introduces a novel generative model for creating layouts by constructing a constraint graph with transformer-based element generation and efficient constrained optimization, enabling high-quality, user-input-free, and conditionally adaptable layouts.

Contribution

It presents a new three-step layout generation process combining transformer architectures with constraint optimization, advancing the state of the art in automatic layout design.

Findings

Requires no user input for layout generation

Produces higher quality layouts than previous methods

Enables versatile conditional layout generation

Abstract

We propose a new generative model for layout generation. We generate layouts in three steps. First, we generate the layout elements as nodes in a layout graph. Second, we compute constraints between layout elements as edges in the layout graph. Third, we solve for the final layout using constrained optimization. For the first two steps, we build on recent transformer architectures. The layout optimization implements the constraints efficiently. We show three practical contributions compared to the state of the art: our work requires no user input, produces higher quality layouts, and enables many novel capabilities for conditional layout generation.