SWAGAN: A Style-based Wavelet-driven Generative Model

TL;DR

SWAGAN introduces a wavelet-based, frequency-aware GAN architecture that enhances high-frequency content quality and computational efficiency in image generation, building upon StyleGAN2 for improved realism and editing capabilities.

Contribution

It presents a novel wavelet-driven, style-based GAN architecture that improves high-frequency detail and computational performance in image synthesis.

Findings

Enhanced high-frequency content in generated images.

Improved computational efficiency over traditional GANs.

Maintained editing capabilities of StyleGAN2.

Abstract

In recent years, considerable progress has been made in the visual quality of Generative Adversarial Networks (GANs). Even so, these networks still suffer from degradation in quality for high-frequency content, stemming from a spectrally biased architecture, and similarly unfavorable loss functions. To address this issue, we present a novel general-purpose Style and WAvelet based GAN (SWAGAN) that implements progressive generation in the frequency domain. SWAGAN incorporates wavelets throughout its generator and discriminator architectures, enforcing a frequency-aware latent representation at every step of the way. This approach yields enhancements in the visual quality of the generated images, and considerably increases computational performance. We demonstrate the advantage of our method by integrating it into the SyleGAN2 framework, and verifying that content generation in the wavelet domain leads to higher quality images with more realistic high-frequency content. Furthermore, we verify that our model's latent space retains the qualities that allow StyleGAN to serve as a basis for a multitude of editing tasks, and show that our frequency-aware approach also induces improved downstream visual quality.