Fine-Grained Control of Artistic Styles in Image Generation

TL;DR

This paper introduces a method to embed artworks into a continuous style space, enabling fine-grained control over artistic styles in image generation while improving image quality.

Contribution

It proposes a novel style embedding approach that captures the continuous spectrum of artistic styles and integrates with existing GANs like StyleGAN.

Findings

Precise control over fine-grained artistic styles.

Improved image quality measured by FID.

Compatible with common GAN architectures.

Abstract

Recent advances in generative models and adversarial training have enabled artificially generating artworks in various artistic styles. It is highly desirable to gain more control over the generated style in practice. However, artistic styles are unlike object categories -- there are a continuous spectrum of styles distinguished by subtle differences. Few works have been explored to capture the continuous spectrum of styles and apply it to a style generation task. In this paper, we propose to achieve this by embedding original artwork examples into a continuous style space. The style vectors are fed to the generator and discriminator to achieve fine-grained control. Our method can be used with common generative adversarial networks (such as StyleGAN). Experiments show that our method not only precisely controls the fine-grained artistic style but also improves image quality over vanilla StyleGAN as measured by FID.