MontageGAN: Generation and Assembly of Multiple Components by GANs

TL;DR

MontageGAN is a novel GAN framework that generates multi-layer images by combining local GANs for individual layers and a global GAN for their placement, advancing multi-layer image synthesis.

Contribution

The paper introduces MontageGAN, a new two-step GAN approach for generating and assembling multi-layer images, addressing a gap in existing single-layer focused methods.

Findings

Successfully generates multi-layer images with plausible layer placement

Demonstrates the effectiveness of local and global GAN components

Achieves promising results in layer assembly and image realism

Abstract

A multi-layer image is more valuable than a single-layer image from a graphic designer's perspective. However, most of the proposed image generation methods so far focus on single-layer images. In this paper, we propose MontageGAN, which is a Generative Adversarial Networks (GAN) framework for generating multi-layer images. Our method utilized a two-step approach consisting of local GANs and global GAN. Each local GAN learns to generate a specific image layer, and the global GAN learns the placement of each generated image layer. Through our experiments, we show the ability of our method to generate multi-layer images and estimate the placement of the generated image layers.