A Style-Based Generator Architecture for Generative Adversarial Networks

Published in

Computer Vision and Pattern Recognition(2018)

External Links:

Generate Graph DownloadPDF

TL;DR

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.

Abstract

We propose an alternative generator architecture for generative adversarial networks, borrowing from style transfer literature. The new architecture leads to an automatically learned, unsupervised separation of high-level attributes (e.g., pose and identity when trained on human faces) and stochastic variation in the generated images (e.g., freckles, hair), and it enables intuitive, scale-specific control of the synthesis. The new generator 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. To quantify interpolation quality and disentanglement, we propose two new, automated methods that are applicable to any generator architecture. Finally, we introduce a new, highly varied and high-quality dataset of human faces.

Authors

S. Laine

8 papers

Timo Aila

8 papers

Tero Karras

5 papers

References64 items

Making Convolutional Networks Shift-Invariant Again

Large Scale GAN Training for High Fidelity Natural Image Synthesis

On Self Modulation for Generative Adversarial Networks

GAN Dissection: Visualizing and Understanding Generative Adversarial Networks

Gaussian Mixture Generative Adversarial Networks for Diverse Datasets, and the Unsupervised Clustering of Images

A Style-Based Generator Architecture for Generative Adversarial Networks

Published in

Computer Vision and Pattern Recognition(2018)

External Links:

Generate Graph DownloadPDF

TL;DR

Abstract

Authors

S. Laine

8 papers

Timo Aila

8 papers

Tero Karras

5 papers

References64 items

Making Convolutional Networks Shift-Invariant Again

Large Scale GAN Training for High Fidelity Natural Image Synthesis

On Self Modulation for Generative Adversarial Networks

GAN Dissection: Visualizing and Understanding Generative Adversarial Networks

Gaussian Mixture Generative Adversarial Networks for Diverse Datasets, and the Unsupervised Clustering of Images

Online Adaptative Curriculum Learning for GANs

Dropout-GAN: Learning from a Dynamic Ensemble of Discriminators

Improved Training with Curriculum GANs

Generative adversarial interpolative autoencoding: adversarial training on latent space interpolations encourage convex latent distributions

Glow: Generative Flow with Invertible 1x1 Convolutions

Feature-wise transformations

MIXGAN: Learning Concepts from Different Domains for Mixture Generation

The GAN Landscape: Losses, Architectures, Regularization, and Normalization

Whitening and Coloring transform for GANs

Self-Attention Generative Adversarial Networks

Multimodal Unsupervised Image-to-Image Translation

Disentangling by Factorising

Spectral Normalization for Generative Adversarial Networks

cGANs with Projection Discriminator

A Framework for the Quantitative Evaluation of Disentangled Representations

Isolating Sources of Disentanglement in Variational Autoencoders

Feature-Based Metrics for Exploring the Latent Space of Generative Models

Which Training Methods for GANs do actually Converge?

The Unreasonable Effectiveness of Deep Features as a Perceptual Metric

High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs

Are GANs Created Equal? A Large-Scale Study

Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing their Input Gradients

Progressive Growing of GANs for Improved Quality, Stability, and Variation

GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium

Megapixel Size Image Creation using Generative Adversarial Networks

Universal Style Transfer via Feature Transforms

Exploring the structure of a real-time, arbitrary neural artistic stylization network

Improved Training of Wasserstein GANs

Arbitrary Style Transfer in Real-Time with Adaptive Instance Normalization

Demystifying Neural Style Transfer

A Survey of Inductive Biases for Factorial Representation-Learning

Generative Multi-Adversarial Networks

beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework

A Learned Representation For Artistic Style

Image Style Transfer Using Convolutional Neural Networks

InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets

Improved Techniques for Training GANs

Adversarially Learned Inference

Adversarial Feature Learning

Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks

LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Adam: A Method for Stochastic Optimization

Learning to generate chairs with convolutional neural networks

Very Deep Convolutional Networks for Large-Scale Image Recognition

One millisecond face alignment with an ensemble of regression trees

Stochastic Backpropagation and Approximate Inference in Deep Generative Models

Auto-Encoding Variational Bayes

Disentangling Factors of Variation via Generative Entangling

Improving generalization performance using double backpropagation

Learning Factorial Codes by Predictability Minimization

Animating rotation with quaternion curves

Emergence of invariance and disentangling in deep representations

dsprites: Disentanglement testing sprites dataset

Sampling Generative Networks Notes on a Few Effective Techniques

Dropout: a simple way to prevent neural networks from overfitting

Rectifier Nonlinearities Improve Neural Network Acoustic Models

This Paper Is Included in the Proceedings of the 12th Usenix Symposium on Operating Systems Design and Implementation (osdi '16). Tensorflow: a System for Large-scale Machine Learning Tensorflow: a System for Large-scale Machine Learning

Field of Study

Computer ScienceMathematics

Journal Information

Name

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

Page

5686-5696

Venue Information

Name

Computer Vision and Pattern Recognition

Type

conference

URL

https://ieeexplore.ieee.org/xpl/conhome.jsp?punumber=1000147

Alternate Names

CVPR
Comput Vis Pattern Recognit

TL;DR

Abstract

Authors

References64 items

Making Convolutional Networks Shift-Invariant Again

Large Scale GAN Training for High Fidelity Natural Image Synthesis

On Self Modulation for Generative Adversarial Networks

GAN Dissection: Visualizing and Understanding Generative Adversarial Networks

Gaussian Mixture Generative Adversarial Networks for Diverse Datasets, and the Unsupervised Clustering of Images

TL;DR

Abstract

Authors

References64 items

Making Convolutional Networks Shift-Invariant Again

Large Scale GAN Training for High Fidelity Natural Image Synthesis

On Self Modulation for Generative Adversarial Networks

GAN Dissection: Visualizing and Understanding Generative Adversarial Networks

Gaussian Mixture Generative Adversarial Networks for Diverse Datasets, and the Unsupervised Clustering of Images

Online Adaptative Curriculum Learning for GANs

Dropout-GAN: Learning from a Dynamic Ensemble of Discriminators

Improved Training with Curriculum GANs

Generative adversarial interpolative autoencoding: adversarial training on latent space interpolations encourage convex latent distributions

Glow: Generative Flow with Invertible 1x1 Convolutions

Feature-wise transformations

MIXGAN: Learning Concepts from Different Domains for Mixture Generation

The GAN Landscape: Losses, Architectures, Regularization, and Normalization

Whitening and Coloring transform for GANs

Self-Attention Generative Adversarial Networks

Multimodal Unsupervised Image-to-Image Translation

Disentangling by Factorising

Spectral Normalization for Generative Adversarial Networks

cGANs with Projection Discriminator

A Framework for the Quantitative Evaluation of Disentangled Representations

Isolating Sources of Disentanglement in Variational Autoencoders

Feature-Based Metrics for Exploring the Latent Space of Generative Models

Which Training Methods for GANs do actually Converge?

The Unreasonable Effectiveness of Deep Features as a Perceptual Metric

High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs

Are GANs Created Equal? A Large-Scale Study

Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing their Input Gradients

Progressive Growing of GANs for Improved Quality, Stability, and Variation

GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium

Megapixel Size Image Creation using Generative Adversarial Networks

Universal Style Transfer via Feature Transforms

Exploring the structure of a real-time, arbitrary neural artistic stylization network

Improved Training of Wasserstein GANs

Arbitrary Style Transfer in Real-Time with Adaptive Instance Normalization

Demystifying Neural Style Transfer

A Survey of Inductive Biases for Factorial Representation-Learning

Generative Multi-Adversarial Networks

beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework

A Learned Representation For Artistic Style

Image Style Transfer Using Convolutional Neural Networks

InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets

Improved Techniques for Training GANs

Adversarially Learned Inference

Adversarial Feature Learning

Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks

LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Adam: A Method for Stochastic Optimization

Learning to generate chairs with convolutional neural networks

Very Deep Convolutional Networks for Large-Scale Image Recognition

One millisecond face alignment with an ensemble of regression trees

Stochastic Backpropagation and Approximate Inference in Deep Generative Models

Auto-Encoding Variational Bayes

Disentangling Factors of Variation via Generative Entangling

Improving generalization performance using double backpropagation

Learning Factorial Codes by Predictability Minimization

Animating rotation with quaternion curves

Hong-Xing Yu

Emergence of invariance and disentangling in deep representations

dsprites: Disentanglement testing sprites dataset

Sampling Generative Networks Notes on a Few Effective Techniques

Dropout: a simple way to prevent neural networks from overfitting

Rectifier Nonlinearities Improve Neural Network Acoustic Models

This Paper Is Included in the Proceedings of the 12th Usenix Symposium on Operating Systems Design and Implementation (osdi '16). Tensorflow: a System for Large-scale Machine Learning Tensorflow: a System for Large-scale Machine Learning

Field of Study

Journal Information

Name