Graph2Pix: A Graph-Based Image to Image Translation Framework

TL;DR

Graph2Pix introduces a novel graph-based image translation framework that leverages historical data structures from Artbreeder to generate high-quality images, outperforming existing methods on benchmarks and human perception.

Contribution

The paper presents a new graph-to-image translation model, Graph2Pix, utilizing tree-like data structures from Artbreeder for improved image synthesis.

Findings

25% improvement in LPIPS metric

81.5% user preference in perception studies

Outperforms several existing image translation frameworks

Abstract

In this paper, we propose a graph-based image-to-image translation framework for generating images. We use rich data collected from the popular creativity platform Artbreeder (http://artbreeder.com), where users interpolate multiple GAN-generated images to create artworks. This unique approach of creating new images leads to a tree-like structure where one can track historical data about the creation of a particular image. Inspired by this structure, we propose a novel graph-to-image translation model called Graph2Pix, which takes a graph and corresponding images as input and generates a single image as output. Our experiments show that Graph2Pix is able to outperform several image-to-image translation frameworks on benchmark metrics, including LPIPS (with a 25% improvement) and human perception studies (n=60), where users preferred the images generated by our method 81.5% of the time. Our source code and dataset are publicly available at https://github.com/catlab-team/graph2pix.