(Clicking on the image will restart the animation)

We show that Flow Matching produces smoother and directed trajectories during the layout generation process compared to diffusion models, significantly speeding up inference and overall improving performance.

Abstract

Finding a suitable layout represents a crucial task for diverse applications in graphic design. Motivated by simpler and smoother sampling trajectories, we explore the use of Flow Matching as an alternative to current diffusion-based layout generation models. Specifically, we propose LayoutFlow, an efficient flow-based model capable of generating high-quality layouts. Instead of progressively denoising the elements of a noisy layout, our method learns to gradually move, or flow, the elements of an initial sample until it reaches its final prediction. In addition, we employ a conditioning scheme that allows us to handle various generation tasks with varying degrees of conditioning with a single model. Empirically, LayoutFlow performs on par with state-of-the-art models while being significantly faster.

Overview

Results

Citation

Acknowledgements

We thank the Mip-NeRF 360 website for providing the template.