Adversarial Latent Autoencoders

@article{Pidhorskyi2020AdversarialLA,
  title={Adversarial Latent Autoencoders},
  author={Stanislav Pidhorskyi and Donald A. Adjeroh and Gianfranco Doretto},
  journal={2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2020},
  pages={14092-14101}
}
Autoencoder networks are unsupervised approaches aiming at combining generative and representational properties by learning simultaneously an encoder-generator map. Although studied extensively, the issues of whether they have the same generative power of GANs, or learn disentangled representations, have not been fully addressed. We introduce an autoencoder that tackles these issues jointly, which we call Adversarial Latent Autoencoder (ALAE). It is a general architecture that can leverage… 
One Model to Reconstruct Them All: A Novel Way to Use the Stochastic Noise in StyleGAN
TLDR
A novel StyleGAN-based autoencoder architecture, which can reconstruct images with very high quality across several data domains, and demonstrates a previously unknown grade of generalizablility by training the encoder and decoder independently and on different datasets.
FlexAE: flexibly learning latent priors for wasserstein auto-encoders
TLDR
FlexAE-SR is the new state-of-the-art for the AE based generative models in terms of generation quality as measured by several metrics such as Fréchet Inception Distance, Precision/Recall score and a latent-space interpolation based smoothing scheme to address the nonsmoothness that may arise from highly flexible priors.
Autoencoding Video Latents for Adversarial Video Generation
TLDR
AVLAE (Adversarial Video Latent AutoEncoder) is presented which is a two stream latent autoencoder where the video distribution is learned by adversarial training and is proposed to autoencode the motion and appearance latent vectors of the video generator in the adversarial setting.
Mind2Mind : transfer learning for GANs
TLDR
A transfer learning technique for GANs that significantly reduces training time by freezing the low-level layers of both the critic and generator of the original GAN, and proves rigorously, within the framework of optimal transport, a theorem ensuring the convergence of the learning of the transferred GAN.
ReStyle: A Residual-Based StyleGAN Encoder via Iterative Refinement
TLDR
This work presents a novel inversion scheme that extends current encoder-based inversion methods by introducing an iterative refinement mechanism, and presents a residual-based encoder, named ReStyle, which attains improved accuracy compared to current state-of-the-art encoders with a negligible increase in inference time.
Generative Model without Prior Distribution Matching
TLDR
This work proposes to let the prior match the embedding distribution rather than imposing the latent variables to fit the prior, which alleviates the contradiction between topological properties' preserving of data manifold and distribution matching in latent space.
AE-StyleGAN: Improved Training of Style-Based Auto-Encoders
TLDR
This paper describes a new methodology to train a style-based autoencoder where the encoder and generator are optimized end-to-end and shows that the proposed model consistently outperforms baselines in terms of image inversion and generation quality.
GAN Cocktail: mixing GANs without dataset access
TLDR
This work tackling the problem of model merging, given two constraints that often come up in the real world: no access to the original training data, and without increasing the size of the neural network, with a novel, two-stage solution.
A Contrastive Learning Approach for Training Variational Autoencoder Priors
TLDR
Noise contrastive priors are proposed that improve the generative performance of state-of-the-art VAEs by a large margin on the MNIST, CIFAR-10, CelebA 64, and CelebA HQ 256 datasets.
Dual Contradistinctive Generative Autoencoder
TLDR
The two contradistinctive losses in VAE work harmoniously in DC-VAE leading to a significant qualitative and quantitative performance enhancement over the baseline VAEs without architectural changes.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 56 REFERENCES
Adversarial Autoencoders
TLDR
This paper shows how the adversarial autoencoder can be used in applications such as semi-supervised classification, disentangling style and content of images, unsupervised clustering, dimensionality reduction and data visualization, and performed experiments on MNIST, Street View House Numbers and Toronto Face datasets.
Pioneer Networks: Progressively Growing Generative Autoencoder
TLDR
A novel generative autoencoder network model that learns to encode and reconstruct images with high quality and resolution, and supports smooth random sampling from the latent space of the encoder, and merge recent techniques for progressively building up the parts of the network with the recently introduced adversarial encoder–generator network.
VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learning
TLDR
VEEGAN is introduced, which features a reconstructor network, reversing the action of the generator by mapping from data to noise, and resists mode collapsing to a far greater extent than other recent GAN variants, and produces more realistic samples.
It Takes (Only) Two: Adversarial Generator-Encoder Networks
We present a new autoencoder-type architecture that is trainable in an unsupervised mode, sustains both generation and inference, and has the quality of conditional and unconditional samples boosted
The GAN Landscape: Losses, Architectures, Regularization, and Normalization
TLDR
This work reproduces the current state of the art of GANs from a practical perspective, discusses common pitfalls and reproducibility issues, and goes beyond fairly exploring the GAN landscape.
Towards Photographic Image Manipulation with Balanced Growing of Generative Autoencoders
TLDR
A generative autoencoder that provides fast encoding, faithful reconstructions, sharp generated/reconstructed samples in high resolutions, and a well-structured latent space that supports semantic manipulation of the inputs is presented.
A Style-Based Generator Architecture for Generative Adversarial Networks
TLDR
An alternative generator architecture for generative adversarial networks is proposed, borrowing from style transfer literature, that improves the state-of-the-art in terms of traditional distribution quality metrics, leads to demonstrably better interpolation properties, and also better disentangles the latent factors of variation.
Stabilizing Training of Generative Adversarial Networks through Regularization
TLDR
This work proposes a new regularization approach with low computational cost that yields a stable GAN training procedure and demonstrates the effectiveness of this regularizer accross several architectures trained on common benchmark image generation tasks.
VAE with a VampPrior
TLDR
This paper proposes to extend the variational auto-encoder (VAE) framework with a new type of prior called "Variational Mixture of Posteriors" prior, or VampPrior for short, which consists of a mixture distribution with components given by variational posteriors conditioned on learnable pseudo-inputs.
IntroVAE: Introspective Variational Autoencoders for Photographic Image Synthesis
TLDR
A novel introspective variational autoencoder (IntroVAE) model for synthesizing high-resolution photographic images that is capable of self-evaluating the quality of its generated samples and improving itself accordingly and requires no extra discriminators.
...
1
2
3
4
5
...