SWAGAN: A Style-based Wavelet-driven Generative Model
arXiv:2102.06108
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.
References in corpus (6)
- Conditional Generative Adversarial Nets
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
- On the Spectral Bias of Neural Networks
- Fourier Spectrum Discrepancies in Deep Network Generated Images
- SSD-GAN: Measuring the Realness in the Spatial and Spectral Domains
- Image Super-Resolution Using a Wavelet-based Generative Adversarial Network
Cited by in corpus (6)
- Alias-Free Generative Adversarial Networks
- Focal Frequency Loss for Image Reconstruction and Synthesis
- Learning multi-scale local conditional probability models of images
- EGAIN: Extended GAn INversion
- StyleFusion: A Generative Model for Disentangling Spatial Segments
- On the Frequency Bias of Generative Models