AE-StyleGAN: Improved Training of Style-Based Auto-Encoders
arXiv:2110.08718
Abstract
StyleGANs have shown impressive results on data generation and manipulation in recent years, thanks to its disentangled style latent space. A lot of efforts have been made in inverting a pretrained generator, where an encoder is trained ad hoc after the generator is trained in a two-stage fashion. In this paper, we focus on style-based generators asking a scientific question: Does forcing such a generator to reconstruct real data lead to more disentangled latent space and make the inversion process from image to latent space easy? We describe a new methodology to train a style-based autoencoder where the encoder and generator are optimized end-to-end. We show that our proposed model consistently outperforms baselines in terms of image inversion and generation quality. Supplementary, code, and pretrained models are available on the project website.
arXiv commentAccepted at WACV-22
References in corpus (13)
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- Conditional Generative Adversarial Nets
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
- VoxCeleb2: Deep Speaker Recognition
- Adversarial Feature Learning
- Adversarially Learned Inference
- Invertible Conditional GANs for image editing
- Multimodal Unsupervised Image-to-Image Translation
- Interpreting the Latent Space of GANs for Semantic Face Editing
- CR-GAN: Learning Complete Representations for Multi-view Generation
- Unbiased Auxiliary Classifier GANs with MINE
- Editing in Style: Uncovering the Local Semantics of GANs
- Autoencoding Generative Adversarial Networks