BEGAN: Boundary Equilibrium Generative Adversarial Networks

Abstract

We propose a new equilibrium enforcing method paired with a loss derived from the Wasserstein distance for training auto-encoder based Generative Adversarial Networks. This method balances the generator and discriminator during training. Additionally, it provides a new approximate convergence measure, fast and stable training and high visual quality. We also derive a way of controlling the trade-off between image diversity and visual quality. We focus on the image generation task, setting a new milestone in visual quality, even at higher resolutions. This is achieved while using a relatively simple model architecture and a standard training procedure.

References in corpus (3)

• NIPS 2016 Tutorial: Generative Adversarial Networks
• Energy-based Generative Adversarial Network
• Improved generator objectives for GANs

Cited by in corpus (303)

• Generative Adversarial Network in Medical Imaging: A Review
• Progressive Growing of GANs for Improved Quality, Stability, and Variation
• Improved Training of Wasserstein GANs
• Deep Generative Design: Integration of Topology Optimization and Generative Models
• MMD GAN: Towards Deeper Understanding of Moment Matching Network
• U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
• AMP: Adversarial Motion Priors for Stylized Physics-Based Character Control
• How Generative Adversarial Networks and Their Variants Work: An Overview
• Adversarial Learning for Semi-Supervised Semantic Segmentation
• Differentially Private Generative Adversarial Network
• Multimodal Unsupervised Image-to-Image Translation
• Generative Adversarial Networks in Computer Vision: A Survey and Taxonomy
• A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
• Variational Approaches for Auto-Encoding Generative Adversarial Networks
• GANSynth: Adversarial Neural Audio Synthesis
• Generating Adversarial Examples with Adversarial Networks
• On the Reconstruction of Face Images from Deep Face Templates
• Synthesizing Tabular Data using Generative Adversarial Networks
• NetGAN: Generating Graphs via Random Walks
• Wasserstein Auto-Encoders
• Unsupervised Person Re-identification by Deep Asymmetric Metric Embedding
• Demystifying MMD GANs
• Representation Learning by Rotating Your Faces
• An Introduction to Image Synthesis with Generative Adversarial Nets
• StegNet: Mega Image Steganography Capacity with Deep Convolutional Network
• A Comprehensive Overview and Comparative Analysis on Deep Learning Models: CNN, RNN, LSTM, GRU
• Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow
• Improved Techniques for Training Score-Based Generative Models
• Dual Discriminator Generative Adversarial Nets
• Structure-Preserving Image Super-Resolution
• The Robust Manifold Defense: Adversarial Training using Generative Models
• Deep unsupervised learning of turbulence for inflow generation at various Reynolds numbers
• Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis
• Enhanced Balancing GAN: Minority-class Image Generation
• Semi-supervised Learning with GANs: Manifold Invariance with Improved Inference
• RUDDER: Return Decomposition for Delayed Rewards
• Generative Models for Automatic Chemical Design
• Segmentation of Lungs in Chest X-Ray Image Using Generative Adversarial Networks
• Diagnosing and Enhancing VAE Models
• Generative Adversarial Trainer: Defense to Adversarial Perturbations with GAN
• SteganoGAN: High Capacity Image Steganography with GANs
• On Accurate Evaluation of GANs for Language Generation
• Interpreting the Latent Space of GANs for Semantic Face Editing
• A Tutorial on Deep Learning for Music Information Retrieval
• On the Effects of Batch and Weight Normalization in Generative Adversarial Networks
• Robust Imitation of Diverse Behaviors
• Stabilizing Generative Adversarial Networks: A Survey
• Unsupervised Person Re-identification by Soft Multilabel Learning
• FrankenGAN: Guided Detail Synthesis for Building Mass-Models Using Style-Synchonized GANs
• Physics-Informed Generative Adversarial Networks for Stochastic Differential Equations
• Sliced-Wasserstein Autoencoder: An Embarrassingly Simple Generative Model
• DP-GAN: Diversity-Promoting Generative Adversarial Network for Generating Informative and Diversified Text
• Translating and Segmenting Multimodal Medical Volumes with Cycle- and Shape-Consistency Generative Adversarial Network
• On gradient regularizers for MMD GANs
• Multi-Generator Generative Adversarial Nets
• Photographic Text-to-Image Synthesis with a Hierarchically-nested Adversarial Network
• Improving the Improved Training of Wasserstein GANs: A Consistency Term and Its Dual Effect
• Gradient Estimators for Implicit Models
• Activation Maximization Generative Adversarial Nets
• MeshGAN: Non-linear 3D Morphable Models of Faces
• Generative adversarial interpolative autoencoding: adversarial training on latent space interpolations encourage convex latent distributions
• An Improved Evaluation Framework for Generative Adversarial Networks
• ST-GAN: Spatial Transformer Generative Adversarial Networks for Image Compositing
• Structure-Preserving Super Resolution with Gradient Guidance
• Self Adversarial Training for Human Pose Estimation
• Anomaly Detection with Adversarial Dual Autoencoders
• Understanding GANs: the LQG Setting
• Face Synthesis from Visual Attributes via Sketch using Conditional VAEs and GANs
• Unconditional Audio Generation with Generative Adversarial Networks and Cycle Regularization
• Towards Open-Set Identity Preserving Face Synthesis
• Safer Classification by Synthesis
• DiverGAN: An Efficient and Effective Single-Stage Framework for Diverse Text-to-Image Generation
• Semantically Decomposing the Latent Spaces of Generative Adversarial Networks
• Wasserstein CNN: Learning Invariant Features for NIR-VIS Face Recognition
• Coulomb GANs: Provably Optimal Nash Equilibria via Potential Fields
• FaceFeat-GAN: a Two-Stage Approach for Identity-Preserving Face Synthesis
• A survey on Adversarial Recommender Systems: from Attack/Defense strategies to Generative Adversarial Networks
• Improved Training of Generative Adversarial Networks Using Representative Features
• Counterfactuals uncover the modular structure of deep generative models
• Adversarial Domain Adaptation for Stable Brain-Machine Interfaces
• Global Texture Enhancement for Fake Face Detection in the Wild
• GAMIN: An Adversarial Approach to Black-Box Model Inversion
• Complement Face Forensic Detection and Localization with FacialLandmarks
• Memorization Precedes Generation: Learning Unsupervised GANs with Memory Networks
• McGan: Mean and Covariance Feature Matching GAN
• GANs May Have No Nash Equilibria
• Unpaired Speech Enhancement by Acoustic and Adversarial Supervision for Speech Recognition
• FaceShop: Deep Sketch-based Face Image Editing
• Defending Against Adversarial Attacks by Leveraging an Entire GAN
• Regularization Methods for Generative Adversarial Networks: An Overview of Recent Studies
• Physics Informed Deep Learning for Transport in Porous Media. Buckley Leverett Problem
• Semi-supervised Image Attribute Editing using Generative Adversarial Networks
• Generative Adversarial Networks: A Survey Towards Private and Secure Applications
• High-Quality Face Image SR Using Conditional Generative Adversarial Networks
• Bandwidth Extension on Raw Audio via Generative Adversarial Networks
• Stackelberg GAN: Towards Provable Minimax Equilibrium via Multi-Generator Architectures
• Image-to-Image Translation: Methods and Applications
• Coherent Point Drift Networks: Unsupervised Learning of Non-Rigid Point Set Registration
• CariGAN: Caricature Generation through Weakly Paired Adversarial Learning
• Artificial Generation of Big Data for Improving Image Classification: A Generative Adversarial Network Approach on SAR Data
• TorchGAN: A Flexible Framework for GAN Training and Evaluation
• The Riemannian Geometry of Deep Generative Models
• Score and Lyrics-Free Singing Voice Generation
• Conditional Infilling GANs for Data Augmentation in Mammogram Classification
• Negative Momentum for Improved Game Dynamics
• Deep UL2DL: Channel Knowledge Transfer from Uplink to Downlink
• SRGAN: Training Dataset Matters
• Real or Not Real, that is the Question
• Generative Adversarial Networks (GANs): What it can generate and What it cannot?
• Solving Linear Inverse Problems Using GAN Priors: An Algorithm with Provable Guarantees
• Singing Style Transfer Using Cycle-Consistent Boundary Equilibrium Generative Adversarial Networks
• Exposing GAN-synthesized Faces Using Landmark Locations
• Wasserstein GANs Work Because They Fail (to Approximate the Wasserstein Distance)
• Self-Diagnosing GAN: Diagnosing Underrepresented Samples in Generative Adversarial Networks
• Exploring Bias in GAN-based Data Augmentation for Small Samples
• Unbalanced GANs: Pre-training the Generator of Generative Adversarial Network using Variational Autoencoder
• Quantitatively Evaluating GANs With Divergences Proposed for Training
• UGAN: Untraceable GAN for Multi-Domain Face Translation
• Image Morphing with Perceptual Constraints and STN Alignment
• Disentangling Factors of Variation by Mixing Them
• Composite Functional Gradient Learning of Generative Adversarial Models
• Image Processing Using Multi-Code GAN Prior
• Controllable Generative Adversarial Network
• Versatile Auxiliary Classifier with Generative Adversarial Network (VAC+GAN), Multi Class Scenarios
• Regularizing Generative Adversarial Networks under Limited Data
• Reverse Engineering of Generative Models: Inferring Model Hyperparameters from Generated Images
• Comparing Generative Adversarial Network Techniques for Image Creation and Modification
• GraN-GAN: Piecewise Gradient Normalization for Generative Adversarial Networks
• Dynamic Kernel Distillation for Efficient Pose Estimation in Videos
• Differential Generative Adversarial Networks: Synthesizing Non-linear Facial Variations with Limited Number of Training Data
• Improving GAN Training with Probability Ratio Clipping and Sample Reweighting
• Unsupervised 3D Reconstruction from a Single Image via Adversarial Learning
• PolyGAN: High-Order Polynomial Generators
• The Devil is in the Decoder: Classification, Regression and GANs
• Invariance of Weight Distributions in Rectified MLPs
• Evaluating the Disentanglement of Deep Generative Models through Manifold Topology
• Few-shot Unsupervised Domain Adaptation with Image-to-class Sparse Similarity Encoding
• Semi-supervised Feature-Level Attribute Manipulation for Fashion Image Retrieval
• Annealed Generative Adversarial Networks
• Disentangling Factors of Variation with Cycle-Consistent Variational Auto-Encoders
• TripletGAN: Training Generative Model with Triplet Loss
• Variational Implicit Processes
• Frequency-Aware Reconstruction of Fluid Simulations with Generative Networks
• Shape-from-Mask: A Deep Learning Based Human Body Shape Reconstruction from Binary Mask Images
• Wav2Pix: Speech-conditioned Face Generation using Generative Adversarial Networks
• GAN- vs. JPEG2000 Image Compression for Distributed Automotive Perception: Higher Peak SNR Does Not Mean Better Semantic Segmentation
• Inception Score, Label Smoothing, Gradient Vanishing and -log(D(x)) Alternative
• Energy-Based Sequence GANs for Recommendation and Their Connection to Imitation Learning
• Competing Against Equilibria in Zero-Sum Games with Evolving Payoffs
• Improving GAN Training via Binarized Representation Entropy (BRE) Regularization
• Deep Learning Based Computed Tomography Whys and Wherefores
• Face Translation between Images and Videos using Identity-aware CycleGAN
• MACE: A Flexible Framework for Membership Privacy Estimation in Generative Models
• Boosting Noise Robustness of Acoustic Model via Deep Adversarial Training
• Recent Advances of Image Steganography with Generative Adversarial Networks
• Propagation Channel Modeling by Deep learning Techniques
• In-Domain GAN Inversion for Real Image Editing
• Versatile Auxiliary Regressor with Generative Adversarial network (VAR+GAN)
• Generative Minimization Networks: Training GANs Without Competition
• Face Synthesis with Landmark Points from Generative Adversarial Networks and Inverse Latent Space Mapping
• Unsupervised Discovery of Object Landmarks as Structural Representations
• GestureGAN for Hand Gesture-to-Gesture Translation in the Wild
• Persuasive Faces: Generating Faces in Advertisements
• Detecting Anomalous Faces with 'No Peeking' Autoencoders
• JGAN: A Joint Formulation of GAN for Synthesizing Images and Labels
• DeepMasterPrints: Generating MasterPrints for Dictionary Attacks via Latent Variable Evolution
• Fake face detection via adaptive manipulation traces extraction network
• Decorating Your Own Bedroom: Locally Controlling Image Generation with Generative Adversarial Networks
• Statistical Parametric Speech Synthesis Using Generative Adversarial Networks Under A Multi-task Learning Framework
• Adversarially Approximated Autoencoder for Image Generation and Manipulation
• Generative adversarial networks for generation and classification of physical rehabilitation movement episodes
• Adversarial Colorization Of Icons Based On Structure And Color Conditions
• Application-driven Privacy-preserving Data Publishing with Correlated Attributes
• GANs beyond divergence minimization
• Discriminator Feature-based Inference by Recycling the Discriminator of GANs
• Training VAEs Under Structured Residuals
• Improving Consistency and Correctness of Sequence Inpainting using Semantically Guided Generative Adversarial Network
• Unpaired Photo-to-Caricature Translation on Faces in the Wild
• A Wasserstein GAN model with the total variational regularization
• Image Quality Assessment Techniques Show Improved Training and Evaluation of Autoencoder Generative Adversarial Networks
• Synthetic-Neuroscore: Using A Neuro-AI Interface for Evaluating Generative Adversarial Networks
• A Plug-in Method for Representation Factorization in Connectionist Models
• Poincaré Recurrence, Cycles and Spurious Equilibria in Gradient-Descent-Ascent for Non-Convex Non-Concave Zero-Sum Games
• Generative Modeling with Optimal Transport Maps
• SeismoGen: Seismic Waveform Synthesis Using Generative Adversarial Networks
• CDE-GAN: Cooperative Dual Evolution Based Generative Adversarial Network
• IAN: Combining Generative Adversarial Networks for Imaginative Face Generation
• Deep Learning-based Face Super-Resolution: A Survey
• AMIL: Adversarial Multi Instance Learning for Human Pose Estimation
• Cooperative Training of Fast Thinking Initializer and Slow Thinking Solver for Conditional Learning
• Separating Content from Style Using Adversarial Learning for Recognizing Text in the Wild
• DISSECT: Disentangled Simultaneous Explanations via Concept Traversals
• Capturing Variabilities from Computed Tomography Images with Generative Adversarial Networks
• Do GANs leave artificial fingerprints?
• Training Deep Neural Network in Limited Precision
• NAS-DIP: Learning Deep Image Prior with Neural Architecture Search
• Data Augmentation for Enhancing EEG-based Emotion Recognition with Deep Generative Models
• Segmentation Guided Image-to-Image Translation with Adversarial Networks
• Inclusive GAN: Improving Data and Minority Coverage in Generative Models
• MIXGAN: Learning Concepts from Different Domains for Mixture Generation
• Classify and Generate: Using Classification Latent Space Representations for Image Generations
• Reciprocal Adversarial Learning via Characteristic Functions
• Continual Learning: Tackling Catastrophic Forgetting in Deep Neural Networks with Replay Processes
• Evaluating Generative Adversarial Networks on Explicitly Parameterized Distributions
• Mode matching in GANs through latent space learning and inversion
• Learning Disentangled Representations with Latent Variation Predictability
• Extreme Image Coding via Multiscale Autoencoders With Generative Adversarial Optimization
• Multimodal Image-to-Image Translation via Mutual Information Estimation and Maximization
• Boosting Transferability of Targeted Adversarial Examples via Hierarchical Generative Networks
• FaceShapeGene: A Disentangled Shape Representation for Flexible Face Image Editing
• Versatile Auxiliary Classifier with Generative Adversarial Network (VAC+GAN)
• BasketballGAN: Generating Basketball Play Simulation Through Sketching
• Image2GIF: Generating Cinemagraphs using Recurrent Deep Q-Networks
• Learning Disentangled Representations of Satellite Image Time Series
• Landmark Detection in Low Resolution Faces with Semi-Supervised Learning
• SwapText: Image Based Texts Transfer in Scenes
• Improved BiGAN training with marginal likelihood equalization
• Parametrization of stochastic inputs using generative adversarial networks with application in geology
• Curriculum Learning for Deep Generative Models with Clustering
• A Novel Framework for Selection of GANs for an Application
• Learning the Loss Functions in a Discriminative Space for Video Restoration
• Latent Space Optimal Transport for Generative Models
• Structural Consistency and Controllability for Diverse Colorization
• Adversarial Attack Type I: Cheat Classifiers by Significant Changes
• How good is my GAN?
• Risk Bounds for Unsupervised Cross-Domain Mapping with IPMs
• Independent Encoder for Deep Hierarchical Unsupervised Image-to-Image Translation
• Inferential Wasserstein Generative Adversarial Networks
• GMM-Based Generative Adversarial Encoder Learning
• Expectation Learning for Adaptive Crossmodal Stimuli Association
• A Neuro-AI Interface for Evaluating Generative Adversarial Networks
• Speech-to-Singing Conversion based on Boundary Equilibrium GAN
• CLASS: Cross-Level Attention and Supervision for Salient Objects Detection
• "Best-of-Many-Samples" Distribution Matching
• Unsupervised Multi-Domain Multimodal Image-to-Image Translation with Explicit Domain-Constrained Disentanglement
• Robust GANs against Dishonest Adversaries
• Scalable Fine-grained Generated Image Classification Based on Deep Metric Learning
• Tempered Adversarial Networks
• Asymmetric Distribution Measure for Few-shot Learning
• Learning Generative Models of Structured Signals from Their Superposition Using GANs with Application to Denoising and Demixing
• Alternating Phase Projected Gradient Descent with Generative Priors for Solving Compressive Phase Retrieval
• Auto-Embedding Generative Adversarial Networks for High Resolution Image Synthesis
• Sliced Wasserstein Generative Models
• Joint Learning of Neural Networks via Iterative Reweighted Least Squares
• Provably Convergent Algorithms for Solving Inverse Problems Using Generative Models
• Generator Surgery for Compressed Sensing
• Text-to-image Synthesis via Symmetrical Distillation Networks
• Online Kernel based Generative Adversarial Networks
• Unsupervised Learning of Solutions to Differential Equations with Generative Adversarial Networks
• IVE-GAN: Invariant Encoding Generative Adversarial Networks
• Optimal Transport Based Generative Autoencoders
• Post Selection Inference with Incomplete Maximum Mean Discrepancy Estimator
• Encoder-Powered Generative Adversarial Networks
• Learning to Avoid Errors in GANs by Manipulating Input Spaces
• Weakly supervised discriminative feature learning with state information for person identification
• Synthesis of High-Quality Visible Faces from Polarimetric Thermal Faces using Generative Adversarial Networks
• ChainGAN: A sequential approach to GANs
• Implicit Manifold Learning on Generative Adversarial Networks
• Domain Partitioning Network
• Restricting Greed in Training of Generative Adversarial Network
• Stingray Detection of Aerial Images Using Augmented Training Images Generated by A Conditional Generative Model
• Classification of sparsely labeled spatio-temporal data through semi-supervised adversarial learning
• Wasserstein GAN Can Perform PCA
• Deep learning tackles single-cell analysis A survey of deep learning for scRNA-seq analysis
• Complexity Controlled Generative Adversarial Networks
• Anomaly Detection by One Class Latent Regularized Networks
• Deep Automodulators
• Vehicle Image Generation Going Well with The Surroundings
• Covering the News with (AI) Style
• Adversarial Code Learning for Image Generation
• Reducing the Representation Error of GAN Image Priors Using the Deep Decoder
• Linking Generative Adversarial Learning and Binary Classification
• Multiclass non-Adversarial Image Synthesis, with Application to Classification from Very Small Sample
• A Survey on Adversarial Image Synthesis
• Sym-parameterized Dynamic Inference for Mixed-Domain Image Translation
• Fast Face Image Synthesis with Minimal Training
• Sinusoidal wave generating network based on adversarial learning and its application: synthesizing frog sounds for data augmentation
• Learning to navigate image manifolds induced by generative adversarial networks for unsupervised video generation
• Recognition Awareness: An Application of Latent Cognizance to Open-Set Recognition
• Generating the support with extreme value losses
• Learning Priors for Adversarial Autoencoders
• DefogGAN: Predicting Hidden Information in the StarCraft Fog of War with Generative Adversarial Nets
• Dual Encoder-Decoder based Generative Adversarial Networks for Disentangled Facial Representation Learning
• Normalized Diversification
• Region Growing Curriculum Generation for Reinforcement Learning
• Can NetGAN be improved on short random walks?
• Host-Pathongen Co-evolution Inspired Algorithm Enables Robust GAN Training
• Escaping from Collapsing Modes in a Constrained Space
• A Weak Supervision Approach to Detecting Visual Anomalies for Automated Testing of Graphics Units
• Image Semantic Transformation: Faster, Lighter and Stronger
• Latent Transformations for Object View Points Synthesis
• LEGAN: Disentangled Manipulation of Directional Lighting and Facial Expressions by Leveraging Human Perceptual Judgements
• TiVGAN: Text to Image to Video Generation with Step-by-Step Evolutionary Generator
• Generative Imaging and Image Processing via Generative Encoder
• Video-to-Video Translation for Visual Speech Synthesis
• Dist-GAN: An Improved GAN using Distance Constraints
• Stacked Wasserstein Autoencoder
• High Diversity Attribute Guided Face Generation with GANs
• Solving Inverse Problems with Conditional-GAN Prior via Fast Network-Projected Gradient Descent
• Organ At Risk Segmentation with Multiple Modality
• A study on the use of Boundary Equilibrium GAN for Approximate Frontalization of Unconstrained Faces to aid in Surveillance
• Zero-Shot Learning Based on Knowledge Sharing
• DINO: A Conditional Energy-Based GAN for Domain Translation