Looks Like Magic: Transfer Learning in GANs to Generate New Card Illustrations

TL;DR

This paper introduces MAGICSTYLEGAN, a transfer learning approach using GANs to generate diverse and high-quality Magic: The Gathering card illustrations, demonstrating its effectiveness over simpler models and exploring transfer learning properties.

Contribution

The paper presents MAGICSTYLEGAN models trained on a new diverse dataset for generating Magic card illustrations, showcasing transfer learning capabilities and state-of-the-art results.

Findings

MAGICSTYLEGAN produces high-quality, diverse illustrations.

Transfer learning effectively adapts pre-trained GAN features to new domains.

Simpler models like DCGANs fail to generate proper illustrations.

Abstract

In this paper, we propose MAGICSTYLEGAN and MAGICSTYLEGAN-ADA - both incarnations of the state-of-the-art models StyleGan2 and StyleGan2 ADA - to experiment with their capacity of transfer learning into a rather different domain: creating new illustrations for the vast universe of the game "Magic: The Gathering" cards. This is a challenging task especially due to the variety of elements present in these illustrations, such as humans, creatures, artifacts, and landscapes - not to mention the plethora of art styles of the images made by various artists throughout the years. To solve the task at hand, we introduced a novel dataset, named MTG, with thousands of illustration from diverse card types and rich in metadata. The resulting set is a dataset composed by a myriad of both realistic and fantasy-like illustrations. Although, to investigate effects of diversity we also introduced subsets that contain specific types of concepts, such as forests, islands, faces, and humans. We show that simpler models, such as DCGANs, are not able to learn to generate proper illustrations in any setting. On the other side, we train instances of MAGICSTYLEGAN using all proposed subsets, being able to generate high quality illustrations. We perform experiments to understand how well pre-trained features from StyleGan2 can be transferred towards the target domain. We show that in well trained models we can find particular instances of noise vector that realistically represent real images from the dataset. Moreover, we provide both quantitative and qualitative studies to support our claims, and that demonstrate that MAGICSTYLEGAN is the state-of-the-art approach for generating Magic illustrations. Finally, this paper highlights some emerging properties regarding transfer learning in GANs, which is still a somehow under-explored field in generative learning research.