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

Abstract

Generative Adversarial Networks (GANs) excel at creating realistic images with complex models for which maximum likelihood is infeasible. However, the convergence of GAN training has still not been proved. We propose a two time-scale update rule (TTUR) for training GANs with stochastic gradient descent on arbitrary GAN loss functions. TTUR has an individual learning rate for both the discriminator and the generator. Using the theory of stochastic approximation, we prove that the TTUR converges under mild assumptions to a stationary local Nash equilibrium. The convergence carries over to the popular Adam optimization, for which we prove that it follows the dynamics of a heavy ball with friction and thus prefers flat minima in the objective landscape. For the evaluation of the performance of GANs at image generation, we introduce the "Fréchet Inception Distance" (FID) which captures the similarity of generated images to real ones better than the Inception Score. In experiments, TTUR improves learning for DCGANs and Improved Wasserstein GANs (WGAN-GP) outperforming conventional GAN training on CelebA, CIFAR-10, SVHN, LSUN Bedrooms, and the One Billion Word Benchmark.

Implementations are available at: https://github.com/bioinf-jku/TTUR

Cited by in corpus (1011)

• Diffusion Models Beat GANs on Image Synthesis
• Zero-Shot Text-to-Image Generation
• Alias-Free Generative Adversarial Networks
• Deep learning for molecular design - a review of the state of the art
• ResViT: Residual vision transformers for multi-modal medical image synthesis
• EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning
• Deep Learning for Chest X-ray Analysis: A Survey
• Cascaded Diffusion Models for High Fidelity Image Generation
• Deep Industrial Image Anomaly Detection: A Survey
• Improved Denoising Diffusion Probabilistic Models
• CogView: Mastering Text-to-Image Generation via Transformers
• Deep Learning-enabled Virtual Histological Staining of Biological Samples
• Speech Gesture Generation from the Trimodal Context of Text, Audio, and Speaker Identity
• Deep Learning Approaches for Data Augmentation in Medical Imaging: A Review
• TransGAN: Two Pure Transformers Can Make One Strong GAN, and That Can Scale Up
• Guided Image Generation with Conditional Invertible Neural Networks
• A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
• GANSynth: Adversarial Neural Audio Synthesis
• Adversarial Text-to-Image Synthesis: A Review
• StyleNeRF: A Style-based 3D-Aware Generator for High-resolution Image Synthesis
• Self-supervised Visual Feature Learning with Deep Neural Networks: A Survey
• A Large-scale Study of Representation Learning with the Visual Task Adaptation Benchmark
• Road Segmentation for Remote Sensing Images using Adversarial Spatial Pyramid Networks
• DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models
• Game-Theoretic Multiagent Reinforcement Learning
• Selective Synthetic Augmentation with HistoGAN for Improved Histopathology Image Classification
• Large Scale Image Completion via Co-Modulated Generative Adversarial Networks
• Towards the Automatic Anime Characters Creation with Generative Adversarial Networks
• Generative Adversarial Networks (GANs) in Networking: A Comprehensive Survey & Evaluation
• Generative Adversarial Networks for Spatio-temporal Data: A Survey
• Contrastive Learning for Unpaired Image-to-Image Translation
• SinGAN-Seg: Synthetic training data generation for medical image segmentation
• DiffWave: A Versatile Diffusion Model for Audio Synthesis
• Freeze the Discriminator: a Simple Baseline for Fine-Tuning GANs
• AdaBelief Optimizer: Adapting Stepsizes by the Belief in Observed Gradients
• Deep Learning and Knowledge-Based Methods for Computer Aided Molecular Design -- Toward a Unified Approach: State-of-the-Art and Future Directions
• Image Augmentations for GAN Training
• A survey of synthetic data augmentation methods in computer vision
• InterGen: Diffusion-based Multi-human Motion Generation under Complex Interactions
• Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis
• Applications of Generative Adversarial Networks in Neuroimaging and Clinical Neuroscience
• Generating Diverse High-Fidelity Images with VQ-VAE-2
• High Fidelity Speech Synthesis with Adversarial Networks
• Data Augmentation in High Dimensional Low Sample Size Setting Using a Geometry-Based Variational Autoencoder
• Denoising Diffusion Implicit Models
• A Comprehensive Review of Data-Driven Co-Speech Gesture Generation
• High-Fidelity Image Generation With Fewer Labels
• VTGAN: Semi-supervised Retinal Image Synthesis and Disease Prediction using Vision Transformers
• Beyond Self-attention: External Attention using Two Linear Layers for Visual Tasks
• Structured Denoising Diffusion Models in Discrete State-Spaces
• ViTGAN: Training GANs with Vision Transformers
• ShapeAssembly: Learning to Generate Programs for 3D Shape Structure Synthesis
• Motion Puzzle: Arbitrary Motion Style Transfer by Body Part
• On Finding Local Nash Equilibria (and Only Local Nash Equilibria) in Zero-Sum Games
• Deep Spatial Transformation for Pose-Guided Person Image Generation and Animation
• CIPS-3D: A 3D-Aware Generator of GANs Based on Conditionally-Independent Pixel Synthesis
• Learning to Generate Diverse Dance Motions with Transformer
• LDMVFI: Video Frame Interpolation with Latent Diffusion Models
• Bi-Modality Medical Image Synthesis Using Semi-Supervised Sequential Generative Adversarial Networks
• Deep Generative Modelling: A Comparative Review of VAEs, GANs, Normalizing Flows, Energy-Based and Autoregressive Models
• UNIT-DDPM: UNpaired Image Translation with Denoising Diffusion Probabilistic Models
• You Only Need Adversarial Supervision for Semantic Image Synthesis
• SoftAdapt: Techniques for Adaptive Loss Weighting of Neural Networks with Multi-Part Loss Functions
• A Good Image Generator Is What You Need for High-Resolution Video Synthesis
• Sliced Score Matching: A Scalable Approach to Density and Score Estimation
• Conditional GAN for timeseries generation
• A Systematic Survey of Regularization and Normalization in GANs
• Maximum Entropy Generators for Energy-Based Models
• Knowledge Distillation in Iterative Generative Models for Improved Sampling Speed
• Compressing GANs using Knowledge Distillation
• Diffusion Schrödinger Bridge with Applications to Score-Based Generative Modeling
• Airfoil GAN: Encoding and Synthesizing Airfoils for Aerodynamic Shape Optimization
• Improved StyleGAN Embedding: Where are the Good Latents?
• Neural Actor: Neural Free-view Synthesis of Human Actors with Pose Control
• Task Agnostic Continual Learning via Meta Learning
• Spatial Evolutionary Generative Adversarial Networks
• Lightweight Modules for Efficient Deep Learning based Image Restoration
• i-Mix: A Domain-Agnostic Strategy for Contrastive Representation Learning
• Privacy Protection in Street-View Panoramas using Depth and Multi-View Imagery
• On Fast Sampling of Diffusion Probabilistic Models
• Few-Shot Adaptation of Generative Adversarial Networks
• Is Attention Better Than Matrix Decomposition?
• ImageBART: Bidirectional Context with Multinomial Diffusion for Autoregressive Image Synthesis
• On the Binding Problem in Artificial Neural Networks
• StyleCariGAN: Caricature Generation via StyleGAN Feature Map Modulation
• Learning to Efficiently Sample from Diffusion Probabilistic Models
• Finger-GAN: Generating Realistic Fingerprint Images Using Connectivity Imposed GAN
• Consistency Regularization for Generative Adversarial Networks
• MeshGAN: Non-linear 3D Morphable Models of Faces
• ATISS: Autoregressive Transformers for Indoor Scene Synthesis
• medigan: a Python library of pretrained generative models for medical image synthesis
• IB-GAN: Disentangled Representation Learning with Information Bottleneck Generative Adversarial Networks
• Maximum Likelihood Training of Score-Based Diffusion Models
• Prescribed Generative Adversarial Networks
• Generative adversarial networks in time series: A survey and taxonomy
• Object-driven Text-to-Image Synthesis via Adversarial Training
• DM-GAN: Dynamic Memory Generative Adversarial Networks for Text-to-Image Synthesis
• Time-Travel Rephotography
• DermGAN: Synthetic Generation of Clinical Skin Images with Pathology
• Generative Image Inpainting with Segmentation Confusion Adversarial Training and Contrastive Learning
• Semantic Hierarchy Emerges in Deep Generative Representations for Scene Synthesis
• Direct Speech-to-image Translation
• Convolution with even-sized kernels and symmetric padding
• Mutually improved endoscopic image synthesis and landmark detection in unpaired image-to-image translation
• Lipschitz Generative Adversarial Nets
• Noise Estimation for Generative Diffusion Models
• Everybody Sign Now: Translating Spoken Language to Photo Realistic Sign Language Video
• LAFITE: Towards Language-Free Training for Text-to-Image Generation
• Unifying Multimodal Transformer for Bi-directional Image and Text Generation
• Equilibrium and non-Equilibrium regimes in the learning of Restricted Boltzmann Machines
• Mode Seeking Generative Adversarial Networks for Diverse Image Synthesis
• Habitat-Matterport 3D Dataset (HM3D): 1000 Large-scale 3D Environments for Embodied AI
• Inverse Graphics GAN: Learning to Generate 3D Shapes from Unstructured 2D Data
• Quaternion Generative Adversarial Networks
• Cross-Modal Contrastive Learning for Text-to-Image Generation
• Towards Realistic Visual Dubbing with Heterogeneous Sources
• Generative Adversarial Networks for Image and Video Synthesis: Algorithms and Applications
• E-LPIPS: Robust Perceptual Image Similarity via Random Transformation Ensembles
• LiDAR Sensor modeling and Data augmentation with GANs for Autonomous driving
• Generative Adversarial Networks (GANs): An Overview of Theoretical Model, Evaluation Metrics, and Recent Developments
• Small-GAN: Speeding Up GAN Training Using Core-sets
• AI Choreographer: Music Conditioned 3D Dance Generation with AIST++
• DiverGAN: An Efficient and Effective Single-Stage Framework for Diverse Text-to-Image Generation
• High Perceptual Quality Image Denoising with a Posterior Sampling CGAN
• F3A-GAN: Facial Flow for Face Animation with Generative Adversarial Networks
• MelGAN-VC: Voice Conversion and Audio Style Transfer on arbitrarily long samples using Spectrograms
• TediGAN: Text-Guided Diverse Face Image Generation and Manipulation
• Conditional Generation of Medical Images via Disentangled Adversarial Inference
• Data-Efficient Instance Generation from Instance Discrimination
• PConv: Simple yet Effective Convolutional Layer for Generative Adversarial Network
• Contextual Residual Aggregation for Ultra High-Resolution Image Inpainting
• Using latent space regression to analyze and leverage compositionality in GANs
• Iris-GAN: Learning to Generate Realistic Iris Images Using Convolutional GAN
• Generating Multiple Objects at Spatially Distinct Locations
• Adversarial score matching and improved sampling for image generation
• Focal Frequency Loss for Image Reconstruction and Synthesis
• On Training Sample Memorization: Lessons from Benchmarking Generative Modeling with a Large-scale Competition
• Attention2AngioGAN: Synthesizing Fluorescein Angiography from Retinal Fundus Images using Generative Adversarial Networks
• FedGAN: Federated Generative Adversarial Networks for Distributed Data
• Deceive D: Adaptive Pseudo Augmentation for GAN Training with Limited Data
• Text as Neural Operator: Image Manipulation by Text Instruction
• Improved Contrastive Divergence Training of Energy Based Models
• Physics-informed GANs for Coastal Flood Visualization
• Rethinking Image Inpainting via a Mutual Encoder-Decoder with Feature Equalizations
• Why Spectral Normalization Stabilizes GANs: Analysis and Improvements
• An Overview of Voice Conversion and its Challenges: From Statistical Modeling to Deep Learning
• Complement Face Forensic Detection and Localization with FacialLandmarks
• Image Amodal Completion: A Survey
• Improving Inversion and Generation Diversity in StyleGAN using a Gaussianized Latent Space
• GANs May Have No Nash Equilibria
• GAN-QP: A Novel GAN Framework without Gradient Vanishing and Lipschitz Constraint
• Towards Real-World Blind Face Restoration with Generative Facial Prior
• DeepFlow: History Matching in the Space of Deep Generative Models
• Probabilistic Character Motion Synthesis using a Hierarchical Deep Latent Variable Model
• X-LXMERT: Paint, Caption and Answer Questions with Multi-Modal Transformers
• Cross-domain Correspondence Learning for Exemplar-based Image Translation
• Regularization Methods for Generative Adversarial Networks: An Overview of Recent Studies
• Cross-Domain Few-Shot Learning by Representation Fusion
• Weather GAN: Multi-Domain Weather Translation Using Generative Adversarial Networks
• Improving Text-to-Image Synthesis Using Contrastive Learning
• ShapeMOD: Macro Operation Discovery for 3D Shape Programs
• FCC-GAN: A Fully Connected and Convolutional Net Architecture for GANs
• DP-Image: Differential Privacy for Image Data in Feature Space
• Augmented Normalizing Flows: Bridging the Gap Between Generative Flows and Latent Variable Models
• Cross Modal Compression: Towards Human-comprehensible Semantic Compression
• Neural Language Generation: Formulation, Methods, and Evaluation
• One-Shot Free-View Neural Talking-Head Synthesis for Video Conferencing
• Towards Open-World Text-Guided Face Image Generation and Manipulation
• Universal Rate-Distortion-Perception Representations for Lossy Compression
• How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Models
• Hybrid Discriminative-Generative Training via Contrastive Learning
• Improving the Fairness of Deep Generative Models without Retraining
• GAN Prior Embedded Network for Blind Face Restoration in the Wild
• Text-to-Face Generation with StyleGAN2
• Self-Supervised Variational Auto-Encoders
• Adversarial Learning for Improved Onsets and Frames Music Transcription
• Creative Sketch Generation
• InfinityGAN: Towards Infinite-Pixel Image Synthesis
• Responsible Disclosure of Generative Models Using Scalable Fingerprinting
• Fast and Provable ADMM for Learning with Generative Priors
• High Fidelity Video Prediction with Large Stochastic Recurrent Neural Networks
• Smoothness and Stability in GANs
• Review of end-to-end speech synthesis technology based on deep learning
• House-GAN++: Generative Adversarial Layout Refinement Networks
• Using Scene Graph Context to Improve Image Generation
• A Spectral Energy Distance for Parallel Speech Synthesis
• Encoding Invariances in Deep Generative Models
• From Here to There: Video Inbetweening Using Direct 3D Convolutions
• FairFaceGAN: Fairness-aware Facial Image-to-Image Translation
• Text and Style Conditioned GAN for Generation of Offline Handwriting Lines
• Exploring the Evolution of GANs through Quality Diversity
• Gotta Go Fast When Generating Data with Score-Based Models
• Remote Sensing Image Translation via Style-Based Recalibration Module and Improved Style Discriminator
• LaFIn: Generative Landmark Guided Face Inpainting
• Taming Visually Guided Sound Generation
• Image-to-Image Translation: Methods and Applications
• Conservation AI: Live Stream Analysis for the Detection of Endangered Species Using Convolutional Neural Networks and Drone Technology
• 3DGAUnet: 3D generative adversarial networks with a 3D U-Net based generator to achieve the accurate and effective synthesis of clinical tumor image data for pancreatic cancer
• Mimicry: Towards the Reproducibility of GAN Research
• Enhancing Mechanical Metamodels with a Generative Model-Based Augmented Training Dataset
• On Solving Minimax Optimization Locally: A Follow-the-Ridge Approach
• Towards Realistic 3D Embedding via View Alignment
• Closed-Loop Data Transcription to an LDR via Minimaxing Rate Reduction
• Exploring DeshuffleGANs in Self-Supervised Generative Adversarial Networks
• 4D Facial Expression Diffusion Model
• Age-Oriented Face Synthesis with Conditional Discriminator Pool and Adversarial Triplet Loss
• IR-GAN: Image Manipulation with Linguistic Instruction by Increment Reasoning
• Non Gaussian Denoising Diffusion Models
• InMoDeGAN: Interpretable Motion Decomposition Generative Adversarial Network for Video Generation
• Performance of GAN-based augmentation for deep learning COVID-19 image classification
• Fréchet Audio Distance: A Metric for Evaluating Music Enhancement Algorithms
• Graph Generative Adversarial Networks for Sparse Data Generation in High Energy Physics
• Multi-StyleGAN: Towards Image-Based Simulation of Time-Lapse Live-Cell Microscopy
• Compressing PDF sets using generative adversarial networks
• A Non-Parametric Test to Detect Data-Copying in Generative Models
• Interactive Sketch & Fill: Multiclass Sketch-to-Image Translation
• Guided Image-to-Image Translation with Bi-Directional Feature Transformation
• Disentangled Recurrent Wasserstein Autoencoder
• Learning Energy-Based Models by Diffusion Recovery Likelihood
• Real or Not Real, that is the Question
• Autoencoding Variational Autoencoder
• GENESIS-V2: Inferring Unordered Object Representations without Iterative Refinement
• Dual Contrastive Learning for Unsupervised Image-to-Image Translation
• The Six Fronts of the Generative Adversarial Networks
• Variational Autoencoders with Normalizing Flow Decoders
• Few-shot Image Generation via Cross-domain Correspondence
• Learning stochastic object models from medical imaging measurements by use of advanced ambient generative adversarial networks
• Towards 3D Dance Motion Synthesis and Control
• SRGAN: Training Dataset Matters
• Diverse Image Inpainting with Bidirectional and Autoregressive Transformers
• This Person (Probably) Exists. Identity Membership Attacks Against GAN Generated Faces
• Latent Diffusion Models with Image-Derived Annotations for Enhanced AI-Assisted Cancer Diagnosis in Histopathology
• GramGAN: Deep 3D Texture Synthesis From 2D Exemplars
• Procedural content generation of puzzle games using conditional generative adversarial networks
• A Utility-Preserving GAN for Face Obscuration
• K-Hairstyle: A Large-scale Korean Hairstyle Dataset for Virtual Hair Editing and Hairstyle Classification
• Self-Diagnosing GAN: Diagnosing Underrepresented Samples in Generative Adversarial Networks
• MISS GAN: A Multi-IlluStrator Style Generative Adversarial Network for image to illustration translation
• Bridging the gap between paired and unpaired medical image translation
• Progressive Semantic-Aware Style Transformation for Blind Face Restoration
• Adversarial representation learning for private speech generation
• Rethinking Sampling in 3D Point Cloud Generative Adversarial Networks
• Accelerating Sparse Deep Neural Networks
• Latent Translation: Crossing Modalities by Bridging Generative Models
• Attribute-specific Control Units in StyleGAN for Fine-grained Image Manipulation
• A likelihood approach to nonparametric estimation of a singular distribution using deep generative models
• Video Generation from Single Semantic Label Map
• Improving the Speed and Quality of GAN by Adversarial Training
• Virtual Staining of Label-Free Tissue in Imaging Mass Spectrometry
• Generating unseen complex scenes: are we there yet?
• MobileStyleGAN: A Lightweight Convolutional Neural Network for High-Fidelity Image Synthesis
• SSD-GAN: Measuring the Realness in the Spatial and Spectral Domains
• Composite Score for Anomaly Detection in Imbalanced Real-World Industrial Dataset
• Unbalanced GANs: Pre-training the Generator of Generative Adversarial Network using Variational Autoencoder
• Example-Guided Style Consistent Image Synthesis from Semantic Labeling
• Explanation by Progressive Exaggeration
• Unsupervised Medical Image Segmentation with Adversarial Networks: From Edge Diagrams to Segmentation Maps
• Score-based Generative Modeling in Latent Space
• DeepFaceEditing: Deep Face Generation and Editing with Disentangled Geometry and Appearance Control
• SIGAN: A Novel Image Generation Method for Solar Cell Defect Segmentation and Augmentation
• Image Morphing with Perceptual Constraints and STN Alignment
• FTGAN: A Fully-trained Generative Adversarial Networks for Text to Face Generation
• Generating unrepresented proportions of geological facies using Generative Adversarial Networks
• OPA: Object Placement Assessment Dataset
• Understanding Overparameterization in Generative Adversarial Networks
• Adversarial Generation of Time-Frequency Features with application in audio synthesis
• Diverse Multimedia Layout Generation with Multi Choice Learning
• Style Transfer with Diffusion Models for Synthetic-to-Real Domain Adaptation
• AdvSPADE: Realistic Unrestricted Attacks for Semantic Segmentation
• Large-scale multilingual audio visual dubbing
• Jointly Measuring Diversity and Quality in Text Generation Models
• Lightweight Generative Adversarial Networks for Text-Guided Image Manipulation
• Image Comes Dancing with Collaborative Parsing-Flow Video Synthesis
• Unsupervised Image-to-Image Translation via Pre-trained StyleGAN2 Network
• Implicit Rank-Minimizing Autoencoder
• Self-Adversarial Learning with Comparative Discrimination for Text Generation
• Data Augmentation in Earth Observation: A Diffusion Model Approach
• Feature Unlearning for Pre-trained GANs and VAEs
• FaR-GAN for One-Shot Face Reenactment
• CFA-Net: Controllable Face Anonymization Network with Identity Representation Manipulation
• Unbiased Auxiliary Classifier GANs with MINE
• Generative Convolution Layer for Image Generation
• Parser-Free Virtual Try-on via Distilling Appearance Flows
• More Photos are All You Need: Semi-Supervised Learning for Fine-Grained Sketch Based Image Retrieval
• PISE: Person Image Synthesis and Editing with Decoupled GAN
• Regularizing Generative Adversarial Networks under Limited Data
• Densely connected normalizing flows
• Do Neural Optimal Transport Solvers Work? A Continuous Wasserstein-2 Benchmark
• Investigating Under and Overfitting in Wasserstein Generative Adversarial Networks
• Image Obfuscation for Privacy-Preserving Machine Learning
• Alleviating Mode Collapse in GAN via Diversity Penalty Module
• A Brief Introduction to Generative Models
• Texture Mixer: A Network for Controllable Synthesis and Interpolation of Texture
• Human Motion Transfer from Poses in the Wild
• Free-form Video Inpainting with 3D Gated Convolution and Temporal PatchGAN
• PIRenderer: Controllable Portrait Image Generation via Semantic Neural Rendering
• Realistic Ultrasound Image Synthesis for Improved Classification of Liver Disease
• Generating Diverse Structure for Image Inpainting With Hierarchical VQ-VAE
• PI-REC: Progressive Image Reconstruction Network With Edge and Color Domain
• GraN-GAN: Piecewise Gradient Normalization for Generative Adversarial Networks
• Isometric Autoencoders
• IMUTube: Automatic Extraction of Virtual on-body Accelerometry from Video for Human Activity Recognition
• The Spatially-Correlative Loss for Various Image Translation Tasks
• MulGAN: Facial Attribute Editing by Exemplar
• MixerGAN: An MLP-Based Architecture for Unpaired Image-to-Image Translation
• Deep Exemplar-based Video Colorization
• Joint Generative and Contrastive Learning for Unsupervised Person Re-identification
• Pixel-wise Conditioned Generative Adversarial Networks for Image Synthesis and Completion
• Evaluation in Neural Style Transfer: A Review
• TM-NET: Deep Generative Networks for Textured Meshes
• FIRe-GAN: A novel Deep Learning-based infrared-visible fusion method for wildfire imagery
• EncryptGAN: Image Steganography with Domain Transform
• Mask-Guided Portrait Editing with Conditional GANs
• An Improved Self-supervised GAN via Adversarial Training
• Towards Efficient and Unbiased Implementation of Lipschitz Continuity in GANs
• 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
• (q,p)-Wasserstein GANs: Comparing Ground Metrics for Wasserstein GANs
• Regularized Autoencoders via Relaxed Injective Probability Flow
• Controllable and Progressive Image Extrapolation
• DwNet: Dense warp-based network for pose-guided human video generation
• A Survey and Taxonomy of Adversarial Neural Networks for Text-to-Image Synthesis
• Enhancing the Performance of Practical Profiling Side-Channel Attacks Using Conditional Generative Adversarial Networks
• Fast Mixing of Multi-Scale Langevin Dynamics under the Manifold Hypothesis
• Editing in Style: Uncovering the Local Semantics of GANs
• Imbalanced Data Learning by Minority Class Augmentation using Capsule Adversarial Networks
• Discriminator Contrastive Divergence: Semi-Amortized Generative Modeling by Exploring Energy of the Discriminator
• DG-Font: Deformable Generative Networks for Unsupervised Font Generation
• Generating Images with Sparse Representations
• Image-to-image Translation via Hierarchical Style Disentanglement
• Generative Adversarial U-Net for Domain-free Medical Image Augmentation
• Deep Direct Volume Rendering: Learning Visual Feature Mappings From Exemplary Images
• Audio-Driven Emotional Video Portraits
• Adaptive Model Learning of Neural Networks with UUB Stability for Robot Dynamic Estimation
• PhenDiff: Revealing Subtle Phenotypes with Diffusion Models in Real Images
• LaMD: Latent Motion Diffusion for Image-Conditional Video Generation
• Do Not Escape From the Manifold: Discovering the Local Coordinates on the Latent Space of GANs
• Symbolic Music Generation with Diffusion Models
• Improving Generalization of Transfer Learning Across Domains Using Spatio-Temporal Features in Autonomous Driving
• Effective Universal Unrestricted Adversarial Attacks using a MOE Approach
• Quantitative analysis of robot gesticulation behavior
• A Contrastive Learning Approach for Training Variational Autoencoder Priors
• GAN Slimming: All-in-One GAN Compression by A Unified Optimization Framework
• Relightable 3D Head Portraits from a Smartphone Video
• Continuous Conditional Generative Adversarial Networks: Novel Empirical Losses and Label Input Mechanisms
• Describe What to Change: A Text-guided Unsupervised Image-to-Image Translation Approach
• Zero-Shot Paragraph-level Handwriting Imitation with Latent Diffusion Models
• MCL-GAN: Generative Adversarial Networks with Multiple Specialized Discriminators
• HyperInverter: Improving StyleGAN Inversion via Hypernetwork
• Conditional GANs with Auxiliary Discriminative Classifier
• Dynamics of Fourier Modes in Torus Generative Adversarial Networks
• FLNet: Landmark Driven Fetching and Learning Network for Faithful Talking Facial Animation Synthesis
• Invert and Defend: Model-based Approximate Inversion of Generative Adversarial Networks for Secure Inference
• PROUD: PaRetO-gUided Diffusion Model for Multi-objective Generation
• Human Image Generation: A Comprehensive Survey
• Semantic Bottleneck Scene Generation
• Conditional Image Generation with Score-Based Diffusion Models
• Stylized Face Sketch Extraction via Generative Prior with Limited Data
• COCO-FUNIT: Few-Shot Unsupervised Image Translation with a Content Conditioned Style Encoder
• WaveFill: A Wavelet-based Generation Network for Image Inpainting
• One-Shot Generative Domain Adaptation
• RPGAN: GANs Interpretability via Random Routing
• Learning latent representations across multiple data domains using Lifelong VAEGAN
• Write-a-speaker: Text-based Emotional and Rhythmic Talking-head Generation
• Training GANs with Stronger Augmentations via Contrastive Discriminator
• A Multi-attribute Controllable Generative Model for Histopathology Image Synthesis
• Neural Architecture Search for Deep Image Prior
• FICGAN: Facial Identity Controllable GAN for De-identification
• KoDF: A Large-scale Korean DeepFake Detection Dataset
• Anycost GANs for Interactive Image Synthesis and Editing
• Adversarial Image Composition with Auxiliary Illumination
• Few-shot Video-to-Video Synthesis
• One-Shot Identity-Preserving Portrait Reenactment
• Learning Spatial Pyramid Attentive Pooling in Image Synthesis and Image-to-Image Translation
• Compound Frechet Inception Distance for Quality Assessment of GAN Created Images
• Image Inpainting Guided by Coherence Priors of Semantics and Textures
• Phase Retrieval with Holography and Untrained Priors: Tackling the Challenges of Low-Photon Nanoscale Imaging
• Controllable Person Image Synthesis with Spatially-Adaptive Warped Normalization
• Super-resolution Variational Auto-Encoders
• Positional Encoding as Spatial Inductive Bias in GANs
• Privacy-preserving medical image analysis
• Additional Look into GAN-based Augmentation for Deep Learning COVID-19 Image Classification
• Teachers Do More Than Teach: Compressing Image-to-Image Models
• RankGAN: A Maximum Margin Ranking GAN for Generating Faces
• No MCMC for me: Amortized sampling for fast and stable training of energy-based models
• Language-Driven Image Style Transfer
• $(f,\Gamma )$-Divergences: Interpolating between $f$-Divergences and Integral Probability Metrics
• Training Federated GANs with Theoretical Guarantees: A Universal Aggregation Approach
• AniGAN: Style-Guided Generative Adversarial Networks for Unsupervised Anime Face Generation
• NaturalInversion: Data-Free Image Synthesis Improving Real-World Consistency
• Pix2Next: Leveraging Vision Foundation Models for RGB to NIR Image Translation
• On Data Augmentation and Adversarial Risk: An Empirical Analysis
• Data Separability for Neural Network Classifiers and the Development of a Separability Index
• Trusted Artificial Intelligence: Towards Certification of Machine Learning Applications
• Gradient Descent-Ascent Provably Converges to Strict Local Minmax Equilibria with a Finite Timescale Separation
• Styleformer: Transformer based Generative Adversarial Networks with Style Vector
• A Decentralized Adaptive Momentum Method for Solving a Class of Min-Max Optimization Problems
• Self-Supervised GANs with Label Augmentation
• Neural Architecture Search for Generative Adversarial Networks: A Comprehensive Review and Critical Analysis
• Recapture as You Want
• Generative Minimization Networks: Training GANs Without Competition
• VAEBM: A Symbiosis between Variational Autoencoders and Energy-based Models
• Mosaicking to Distill: Knowledge Distillation from Out-of-Domain Data
• Leveraging Visual Question Answering to Improve Text-to-Image Synthesis
• Convolutional Normalization: Improving Deep Convolutional Network Robustness and Training
• Data-Efficient GAN Training Beyond (Just) Augmentations: A Lottery Ticket Perspective
• Tdcgan: Temporal Dilated Convolutional Generative Adversarial Network for End-to-end Speech Enhancement
• Do We Really Need to Learn Representations from In-domain Data for Outlier Detection?
• Causal Adversarial Network for Learning Conditional and Interventional Distributions
• A Survey of Spatio-Temporal EEG data Analysis: from Models to Applications
• Progressive Limb-Aware Virtual Try-On
• Privacy Leakage of SIFT Features via Deep Generative Model based Image Reconstruction
• Reviewing FID and SID Metrics on Generative Adversarial Networks
• Decentralized Attribution of Generative Models
• Jigsaw-VAE: Towards Balancing Features in Variational Autoencoders
• Single Underwater Image Restoration by Contrastive Learning
• GAN-based Generation and Automatic Selection of Explanations for Neural Networks
• Anytime Sampling for Autoregressive Models via Ordered Autoencoding
• Train simultaneously, generalize better: Stability of gradient-based minimax learners
• Conditional Adversarial Generative Flow for Controllable Image Synthesis
• SceneGen: Learning to Generate Realistic Traffic Scenes
• Roof-GAN: Learning to Generate Roof Geometry and Relations for Residential Houses
• Adversarially Approximated Autoencoder for Image Generation and Manipulation
• Self-labeled Conditional GANs
• Near-optimal Local Convergence of Alternating Gradient Descent-Ascent for Minimax Optimization
• Transposer: Universal Texture Synthesis Using Feature Maps as Transposed Convolution Filter
• SketchyCOCO: Image Generation from Freehand Scene Sketches
• Trumpets: Injective Flows for Inference and Inverse Problems
• Style-based Point Generator with Adversarial Rendering for Point Cloud Completion
• SocialInteractionGAN: Multi-person Interaction Sequence Generation
• Constrained Generative Adversarial Network Ensembles for Sharable Synthetic Data Generation
• CoDeGAN: Contrastive Disentanglement for Generative Adversarial Network
• Unpaired Image Super-Resolution using Pseudo-Supervision
• Denoising Diffusion Gamma Models
• Robust Learning Meets Generative Models: Can Proxy Distributions Improve Adversarial Robustness?
• Information-Theoretic Visual Explanation for Black-Box Classifiers
• Generative Adversarial Zero-Shot Relational Learning for Knowledge Graphs
• Particle Cloud Generation with Message Passing Generative Adversarial Networks
• DeepBlur: A Simple and Effective Method for Natural Image Obfuscation
• TSIT: A Simple and Versatile Framework for Image-to-Image Translation
• Robust Pose Transfer with Dynamic Details using Neural Video Rendering
• Chunked Autoregressive GAN for Conditional Waveform Synthesis
• Likelihood Training of Schrödinger Bridge using Forward-Backward SDEs Theory
• Rethinking Image Deraining via Rain Streaks and Vapors
• Neural Turtle Graphics for Modeling City Road Layouts
• MagGAN: High-Resolution Face Attribute Editing with Mask-Guided Generative Adversarial Network
• Reference-guided Face Component Editing
• Conditional WGANs with Adaptive Gradient Balancing for Sparse MRI Reconstruction
• Efficient Ring-topology Decentralized Federated Learning with Deep Generative Models for Industrial Artificial Intelligent
• Generative Adversarial Networks (GAN) Powered Fast Magnetic Resonance Imaging -- Mini Review, Comparison and Perspectives
• Memory-efficient Learning for High-Dimensional MRI Reconstruction
• Task-agnostic Temporally Consistent Facial Video Editing
• Attentive Normalization for Conditional Image Generation
• PMC-GANs: Generating Multi-Scale High-Quality Pedestrian with Multimodal Cascaded GANs
• Wasserstein-Wasserstein Auto-Encoders
• Simple Video Generation using Neural ODEs
• Diverse Semantic Image Synthesis via Probability Distribution Modeling
• Quality Evaluation of GANs Using Cross Local Intrinsic Dimensionality
• Diffusion-Weighted Magnetic Resonance Brain Images Generation with Generative Adversarial Networks and Variational Autoencoders: A Comparison Study
• Dual Contrastive Loss and Attention for GANs
• GANs Can Play Lottery Tickets Too
• High Resolution Face Editing with Masked GAN Latent Code Optimization
• CoNeS: Conditional neural fields with shift modulation for multi-sequence MRI translation
• Can GAN originate new electronic dance music genres? -- Generating novel rhythm patterns using GAN with Genre Ambiguity Loss
• Using Simulated Data to Generate Images of Climate Change
• ComicGAN: Text-to-Comic Generative Adversarial Network
• Learning Multi-Site Harmonization of Magnetic Resonance Images Without Traveling Human Phantoms
• StyleUV: Diverse and High-fidelity UV Map Generative Model
• Learning by Turning: Neural Architecture Aware Optimisation
• CDE-GAN: Cooperative Dual Evolution Based Generative Adversarial Network
• Bidirectional Mapping Generative Adversarial Networks for Brain MR to PET Synthesis
• Temporally coherent video anonymization through GAN inpainting
• Generative Modeling with Optimal Transport Maps
• Measuring the Biases and Effectiveness of Content-Style Disentanglement
• Using GANs to Synthesise Minimum Training Data for Deepfake Generation
• On Characterizing GAN Convergence Through Proximal Duality Gap
• Natural and Realistic Single Image Super-Resolution with Explicit Natural Manifold Discrimination
• Bi-level Feature Alignment for Versatile Image Translation and Manipulation
• Shape-Guided Clothing Warping for Virtual Try-On
• Learning Neurosymbolic Generative Models via Program Synthesis
• IMAGINE: Image Synthesis by Image-Guided Model Inversion
• FaceController: Controllable Attribute Editing for Face in the Wild
• A survey on Variational Autoencoders from a GreenAI perspective
• FBC-GAN: Diverse and Flexible Image Synthesis via Foreground-Background Composition
• GAN Cocktail: mixing GANs without dataset access
• Deep Learning-based Face Super-Resolution: A Survey
• Hybrid Generative-Contrastive Representation Learning
• Unsupervised Shadow Removal Using Target Consistency Generative Adversarial Network
• Unsupervised Change Detection in Satellite Images with Generative Adversarial Network
• Improving Robustness of Learning-based Autonomous Steering Using Adversarial Images
• Generated Loss and Augmented Training of MNIST VAE
• PEARL: Data Synthesis via Private Embeddings and Adversarial Reconstruction Learning
• BézierSketch: A generative model for scalable vector sketches
• Scalable Balanced Training of Conditional Generative Adversarial Neural Networks on Image Data
• Conditional Image Generation and Manipulation for User-Specified Content
• Coconditional Autoencoding Adversarial Networks for Chinese Font Feature Learning
• CoCosNet v2: Full-Resolution Correspondence Learning for Image Translation
• Biphasic Learning of GANs for High-Resolution Image-to-Image Translation
• Continuous Conditional Generative Adversarial Networks (cGAN) with Generator Regularization
• Dual Attention GANs for Semantic Image Synthesis
• Fast Universal Style Transfer for Artistic and Photorealistic Rendering
• A Game-Theoretic Approach to Multi-Agent Trust Region Optimization
• Example-Guided Image Synthesis across Arbitrary Scenes using Masked Spatial-Channel Attention and Self-Supervision
• Kernel Stein Generative Modeling
• Dual Generator Generative Adversarial Networks for Multi-Domain Image-to-Image Translation
• Deep Generative Learning via Variational Gradient Flow
• CPOT: Channel Pruning via Optimal Transport
• Influence Estimation for Generative Adversarial Networks
• I Want This Product but Different : Multimodal Retrieval with Synthetic Query Expansion
• On Transportation of Mini-batches: A Hierarchical Approach
• MUST-GAN: Multi-level Statistics Transfer for Self-driven Person Image Generation
• DivCo: Diverse Conditional Image Synthesis via Contrastive Generative Adversarial Network
• Disentangling representations of retinal images with generative models
• Score-Guided Generative Adversarial Networks
• Generating Correct Answers for Progressive Matrices Intelligence Tests
• Self-Supervised GAN Compression
• On Quantitative Evaluations of Counterfactuals
• Defending Medical Image Diagnostics against Privacy Attacks using Generative Methods
• Segmentation of Surgical Instruments for Minimally-Invasive Robot-Assisted Procedures Using Generative Deep Neural Networks
• RetrieveGAN: Image Synthesis via Differentiable Patch Retrieval
• MOGAN: Morphologic-structure-aware Generative Learning from a Single Image
• Bridging Global Context Interactions for High-Fidelity Image Completion
• Bayesian modeling of insurance claims for hail damage
• Learning Implicit Generative Models with Theoretical Guarantees
• Talking-head Generation with Rhythmic Head Motion
• Universal Face Restoration With Memorized Modulation
• Learning to Incorporate Structure Knowledge for Image Inpainting
• NTIRE 2021 Challenge on Quality Enhancement of Compressed Video: Methods and Results
• Deep Generation of Face Images from Sketches
• DISSECT: Disentangled Simultaneous Explanations via Concept Traversals
• PowerGAN: Synthesizing Appliance Power Signatures Using Generative Adversarial Networks
• The Art of Food: Meal Image Synthesis from Ingredients
• Selective Synthetic Augmentation with Quality Assurance
• Fast Approximation of the Sliced-Wasserstein Distance Using Concentration of Random Projections
• Generative View Synthesis: From Single-view Semantics to Novel-view Images
• In&Out : Diverse Image Outpainting via GAN Inversion
• Approximating Human Judgment of Generated Image Quality
• Deep Generative Learning via Schrödinger Bridge
• A gradual, semi-discrete approach to generative network training via explicit Wasserstein minimization
• 3D-StyleGAN: A Style-Based Generative Adversarial Network for Generative Modeling of Three-Dimensional Medical Images
• Direct May Not Be the Best: An Incremental Evolution View of Pose Generation
• Manifold Topology Divergence: a Framework for Comparing Data Manifolds
• Automating Generative Deep Learning for Artistic Purposes: Challenges and Opportunities
• DeltaGAN: Towards Diverse Few-shot Image Generation with Sample-Specific Delta
• NAS-DIP: Learning Deep Image Prior with Neural Architecture Search
• Improving Mini-batch Optimal Transport via Partial Transportation
• JOKR: Joint Keypoint Representation for Unsupervised Cross-Domain Motion Retargeting
• Learning Prototype-oriented Set Representations for Meta-Learning
• Learning Semantic Person Image Generation by Region-Adaptive Normalization
• Nonconvex-Nonconcave Min-Max Optimization with a Small Maximization Domain
• Model Extraction and Defenses on Generative Adversarial Networks
• GuidedStyle: Attribute Knowledge Guided Style Manipulation for Semantic Face Editing
• GeoSim: Realistic Video Simulation via Geometry-Aware Composition for Self-Driving
• Attribute2Font: Creating Fonts You Want From Attributes
• Flow Guided Transformable Bottleneck Networks for Motion Retargeting
• Improved Transformer for High-Resolution GANs
• Training Generative Adversarial Networks in One Stage
• DTGAN: Dual Attention Generative Adversarial Networks for Text-to-Image Generation
• Bilateral Asymmetry Guided Counterfactual Generating Network for Mammogram Classification
• PriorGAN: Real Data Prior for Generative Adversarial Nets
• Dual Contradistinctive Generative Autoencoder
• Evaluation Metrics for Graph Generative Models: Problems, Pitfalls, and Practical Solutions
• Adaptive Weighted Discriminator for Training Generative Adversarial Networks
• Bi-level Score Matching for Learning Energy-based Latent Variable Models
• Robustness and Diversity Seeking Data-Free Knowledge Distillation
• Dreaming of Electrical Waves: Generative Modeling of Cardiac Excitation Waves using Diffusion Models
• Cloud2Curve: Generation and Vectorization of Parametric Sketches
• Augmentation-Interpolative AutoEncoders for Unsupervised Few-Shot Image Generation
• ReGO: Reference-Guided Outpainting for Scenery Image
• Image-to-Image Translation with Low Resolution Conditioning
• Unpaired Image-to-Image Translation via Latent Energy Transport
• Example-Guided Scene Image Synthesis using Masked Spatial-Channel Attention and Patch-Based Self-Supervision
• BalaGAN: Image Translation Between Imbalanced Domains via Cross-Modal Transfer
• Deep Fusion Network for Image Completion
• Learning Disentangled Representations with Latent Variation Predictability
• Forward Super-Resolution: How Can GANs Learn Hierarchical Generative Models for Real-World Distributions
• Text-to-Image Generation Grounded by Fine-Grained User Attention
• Domain-Specific Mappings for Generative Adversarial Style Transfer
• Assessing Dialogue Systems with Distribution Distances
• F2GAN: Fusing-and-Filling GAN for Few-shot Image Generation
• Learning Neural Light Transport
• Refining Deep Generative Models via Discriminator Gradient Flow
• SSCR: Iterative Language-Based Image Editing via Self-Supervised Counterfactual Reasoning
• Toward Joint Image Generation and Compression using Generative Adversarial Networks
• Born Identity Network: Multi-way Counterfactual Map Generation to Explain a Classifier's Decision
• Hidden Convexity of Wasserstein GANs: Interpretable Generative Models with Closed-Form Solutions
• MPG: A Multi-ingredient Pizza Image Generator with Conditional StyleGANs
• GANs N' Roses: Stable, Controllable, Diverse Image to Image Translation (works for videos too!)
• Multimodal Image-to-Image Translation via Mutual Information Estimation and Maximization
• Improving Style-Content Disentanglement in Image-to-Image Translation
• Video Reenactment as Inductive Bias for Content-Motion Disentanglement
• CAD-PU: A Curvature-Adaptive Deep Learning Solution for Point Set Upsampling
• Human Pose Transfer with Augmented Disentangled Feature Consistency
• Bringing Old Photos Back to Life
• Less is More: Learning from Synthetic Data with Fine-grained Attributes for Person Re-Identification
• StyleNAS: An Empirical Study of Neural Architecture Search to Uncover Surprisingly Fast End-to-End Universal Style Transfer Networks
• Conditional Denoising of Remote Sensing Imagery Using Cycle-Consistent Deep Generative Models
• Federated CycleGAN for Privacy-Preserving Image-to-Image Translation
• Evaluating Generative Adversarial Networks on Explicitly Parameterized Distributions
• In-N-Out: Towards Good Initialization for Inpainting and Outpainting
• TarGAN: Target-Aware Generative Adversarial Networks for Multi-modality Medical Image Translation
• FaceShapeGene: A Disentangled Shape Representation for Flexible Face Image Editing
• SDA-GAN: Unsupervised Image Translation Using Spectral Domain Attention-Guided Generative Adversarial Network
• Global and Local Alignment Networks for Unpaired Image-to-Image Translation
• Online Exemplar Fine-Tuning for Image-to-Image Translation
• Contrastive Unpaired Translation using Focal Loss for Patch Classification
• Structure-aware Person Image Generation with Pose Decomposition and Semantic Correlation
• Progressively Complementary Network for Fisheye Image Rectification Using Appearance Flow
• Image-to-Image Translation with Multi-Path Consistency Regularization
• M2GAN: A Multi-Stage Self-Attention Network for Image Rain Removal on Autonomous Vehicles
• Hyperbolic Generative Adversarial Network
• MIASSR: An Approach for Medical Image Arbitrary Scale Super-Resolution
• DPD-InfoGAN: Differentially Private Distributed InfoGAN
• Exponential Tilting of Generative Models: Improving Sample Quality by Training and Sampling from Latent Energy
• Boosting Image Outpainting with Semantic Layout Prediction
• Cycle-Consistent Inverse GAN for Text-to-Image Synthesis
• Adversarial Pixel-Level Generation of Semantic Images
• World-Consistent Video-to-Video Synthesis
• Improving the Reconstruction of Disentangled Representation Learners via Multi-Stage Modeling
• Very Long Natural Scenery Image Prediction by Outpainting
• Learning to simulate complex scenes
• CAM-GAN: Continual Adaptation Modules for Generative Adversarial Networks
• PD-GAN: Probabilistic Diverse GAN for Image Inpainting
• Principled Interpolation in Normalizing Flows
• GAN Inversion for Out-of-Range Images with Geometric Transformations
• High-Resolution Complex Scene Synthesis with Transformers
• VICTR: Visual Information Captured Text Representation for Text-to-Image Multimodal Tasks
• UVA: A Universal Variational Framework for Continuous Age Analysis
• Interactive Label Cleaning with Example-based Explanations
• Optimal Transport Relaxations with Application to Wasserstein GANs
• Perceptual Indistinguishability-Net (PI-Net): Facial Image Obfuscation with Manipulable Semantics
• Everything's Talkin': Pareidolia Face Reenactment
• Content-Aware GAN Compression
• Human De-occlusion: Invisible Perception and Recovery for Humans
• Restore from Restored: Single-image Inpainting
• Deep Consensus Learning
• Learning Cycle-Consistent Cooperative Networks via Alternating MCMC Teaching for Unsupervised Cross-Domain Translation
• Protecting Intellectual Property of Generative Adversarial Networks from Ambiguity Attack
• A Generative Model for Hallucinating Diverse Versions of Super Resolution Images
• Generative Learning With Euler Particle Transport
• GMM-Based Generative Adversarial Encoder Learning
• Slimmable Generative Adversarial Networks
• DECOR-GAN: 3D Shape Detailization by Conditional Refinement
• Generating Unobserved Alternatives
• Diffusion models for Handwriting Generation
• Client Adaptation improves Federated Learning with Simulated Non-IID Clients
• Omni-GAN: On the Secrets of cGANs and Beyond
• Vehicle Reconstruction and Texture Estimation Using Deep Implicit Semantic Template Mapping
• PC-GAIN: Pseudo-label Conditional Generative Adversarial Imputation Networks for Incomplete Data
• Improving Augmentation and Evaluation Schemes for Semantic Image Synthesis
• Structure-aware Image Inpainting with Two Parallel Streams
• Manifold-preserved GANs
• Controllable Image Synthesis via SegVAE
• Inferential Wasserstein Generative Adversarial Networks
• Multilevel Knowledge Transfer for Cross-Domain Object Detection
• Artificial Intelligence Hybrid Deep Learning Model for Groundwater Level Prediction Using MLP-ADAM
• InfoVAEGAN : learning joint interpretable representations by information maximization and maximum likelihood
• Independent Encoder for Deep Hierarchical Unsupervised Image-to-Image Translation
• SAFRON: Stitching Across the Frontier for Generating Colorectal Cancer Histology Images
• Unselfie: Translating Selfies to Neutral-pose Portraits in the Wild
• Rectangular Flows for Manifold Learning
• BaMBNet: A Blur-aware Multi-branch Network for Defocus Deblurring
• TinyGAN: Distilling BigGAN for Conditional Image Generation
• Accelerated WGAN update strategy with loss change rate balancing
• Sample weighting as an explanation for mode collapse in generative adversarial networks
• "Best-of-Many-Samples" Distribution Matching
• Exploring Generative Physics Models with Scientific Priors in Inertial Confinement Fusion
• Searching for an (un)stable equilibrium: experiments in training generative models without data
• Transfer Learning with Pre-trained Conditional Generative Models
• Expressive Communication: A Common Framework for Evaluating Developments in Generative Models and Steering Interfaces
• SketchEdit: Mask-Free Local Image Manipulation with Partial Sketches
• Don't Generate Me: Training Differentially Private Generative Models with Sinkhorn Divergence
• On Predicting Generalization using GANs
• DyStyle: Dynamic Neural Network for Multi-Attribute-Conditioned Style Editing
• Harnessing Optoelectronic Noises in a Photonic Generative Network
• Unsupervised Multi-Domain Multimodal Image-to-Image Translation with Explicit Domain-Constrained Disentanglement
• Differential-Critic GAN: Generating What You Want by a Cue of Preferences
• Double Descent and Other Interpolation Phenomena in GANs
• Tractable Density Estimation on Learned Manifolds with Conformal Embedding Flows
• Orthogonal Wasserstein GANs
• CDMA: A Practical Cross-Device Federated Learning Algorithm for General Minimax Problems
• Spherical Image Generation from a Single Normal Field of View Image by Considering Scene Symmetry
• Continuous Face Aging via Self-estimated Residual Age Embedding
• Improving the Evaluation of Generative Models with Fuzzy Logic
• Intelligent Home 3D: Automatic 3D-House Design from Linguistic Descriptions Only
• Exploring The Effect of High-frequency Components in GANs Training
• Lessons Learned from the Training of GANs on Artificial Datasets
• DO-GAN: A Double Oracle Framework for Generative Adversarial Networks
• Deconstructing Generative Adversarial Networks
• Auto-Embedding Generative Adversarial Networks for High Resolution Image Synthesis
• Latent Neural Differential Equations for Video Generation
• not-so-BigGAN: Generating High-Fidelity Images on Small Compute with Wavelet-based Super-Resolution
• Autoregressive Score Matching
• Neural Function Modules with Sparse Arguments: A Dynamic Approach to Integrating Information across Layers
• SMYRF: Efficient Attention using Asymmetric Clustering
• A deep learning based interactive sketching system for fashion images design
• Multivariate-Information Adversarial Ensemble for Scalable Joint Distribution Matching
• Integrating Categorical Semantics into Unsupervised Domain Translation
• Learning Manifold Implicitly via Explicit Heat-Kernel Learning
• BreGMN: scaled-Bregman Generative Modeling Networks
• Intuitive, Interactive Beard and Hair Synthesis with Generative Models
• Unsupervised Multimodal Video-to-Video Translation via Self-Supervised Learning
• Generative Adversarial Data Programming
• Few-shot Compositional Font Generation with Dual Memory
• Semantic Image Manipulation Using Scene Graphs
• A Neuro-AI Interface for Evaluating Generative Adversarial Networks
• Words as Art Materials: Generating Paintings with Sequential GANs
• Robust Conditional GAN from Uncertainty-Aware Pairwise Comparisons
• A Framework and Dataset for Abstract Art Generation via CalligraphyGAN
• Distributional Discrepancy: A Metric for Unconditional Text Generation
• Attributes Aware Face Generation with Generative Adversarial Networks
• Evolutionary Generative Adversarial Networks with Crossover Based Knowledge Distillation
• Detecting GAN generated errors
• A Survey on GAN Acceleration Using Memory Compression Technique
• Inference-InfoGAN: Inference Independence via Embedding Orthogonal Basis Expansion
• Less Memory, Faster Speed: Refining Self-Attention Module for Image Reconstruction
• Exploiting the Hidden Tasks of GANs: Making Implicit Subproblems Explicit
• Deep Image Synthesis from Intuitive User Input: A Review and Perspectives
• A Layer-Based Sequential Framework for Scene Generation with GANs
• Low-Rank Subspaces in GANs
• An Assessment of GANs for Identity-related Applications
• Exploiting Chain Rule and Bayes' Theorem to Compare Probability Distributions
• Semantically Robust Unpaired Image Translation for Data with Unmatched Semantics Statistics
• Self-Supervised Sketch-to-Image Synthesis
• DepthwiseGANs: Fast Training Generative Adversarial Networks for Realistic Image Synthesis
• To Regularize or Not To Regularize? The Bias Variance Trade-off in Regularized AEs
• Correspondence Learning for Controllable Person Image Generation
• SeCGAN: Parallel Conditional Generative Adversarial Networks for Face Editing via Semantic Consistency
• Characterizing Generalization under Out-Of-Distribution Shifts in Deep Metric Learning
• Exploiting Spatial Dimensions of Latent in GAN for Real-time Image Editing
• D2C: Diffusion-Denoising Models for Few-shot Conditional Generation
• CookGAN: Meal Image Synthesis from Ingredients
• On The Distribution of Penultimate Activations of Classification Networks
• Multi-Density Sketch-to-Image Translation Network
• Unsupervised Adversarial Image Inpainting
• GUIGAN: Learning to Generate GUI Designs Using Generative Adversarial Networks
• De-Pois: An Attack-Agnostic Defense against Data Poisoning Attacks
• Realistic Image Synthesis with Configurable 3D Scene Layouts
• L-Verse: Bidirectional Generation Between Image and Text
• IID-GAN: an IID Sampling Perspective for Regularizing Mode Collapse
• Prb-GAN: A Probabilistic Framework for GAN Modelling
• Weakly-supervised Generative Adversarial Networks for medical image classification
• NTIRE 2021 Challenge on Quality Enhancement of Compressed Video: Dataset and Study
• Normalizing Flows with Multi-Scale Autoregressive Priors
• S2cGAN: Semi-Supervised Training of Conditional GANs with Fewer Labels
• Dynamic Neural Garments
• Conditional Frechet Inception Distance
• Adversarial sampling of unknown and high-dimensional conditional distributions
• Toward Accurate and Realistic Outfits Visualization with Attention to Details
• Benefiting Deep Latent Variable Models via Learning the Prior and Removing Latent Regularization
• A Spectral Enabled GAN for Time Series Data Generation
• Semantic Editing On Segmentation Map Via Multi-Expansion Loss
• A New Distributed Method for Training Generative Adversarial Networks
• Variance Constrained Autoencoding
• Noise Homogenization via Multi-Channel Wavelet Filtering for High-Fidelity Sample Generation in GANs
• Random Network Distillation as a Diversity Metric for Both Image and Text Generation
• Implicit Subspace Prior Learning for Dual-Blind Face Restoration
• Conditioning Trick for Training Stable GANs
• Autoencoding Under Normalization Constraints
• Characteristic Regularisation for Super-Resolving Face Images
• Generating Memorable Images Based on Human Visual Memory Schemas
• Optimal Transport Based Generative Autoencoders
• Learning End-to-End Action Interaction by Paired-Embedding Data Augmentation
• Discriminative Cross-Modal Data Augmentation for Medical Imaging Applications
• Generative Transition Mechanism to Image-to-Image Translation via Encoded Transformation
• Interpreting Spatially Infinite Generative Models
• A 3D model-based approach for fitting masks to faces in the wild
• LEED: Label-Free Expression Editing via Disentanglement
• Image Synthesis via Semantic Composition
• Unrealistic Feature Suppression for Generative Adversarial Networks
• Momentum Centering and Asynchronous Update for Adaptive Gradient Methods
• One-Shot Image-to-Image Translation via Part-Global Learning with a Multi-adversarial Framework
• Symmetric Wasserstein Autoencoders
• Neural Crossbreed: Neural Based Image Metamorphosis
• LaDDer: Latent Data Distribution Modelling with a Generative Prior
• Face Attribute Invertion
• Latent Behavior Diffusion for Sequential Reaction Generation in Dyadic Setting
• Person-in-Context Synthesiswith Compositional Structural Space
• TCAM-Diff: Triplane-Aware Cross-Attention Medical Diffusion Model
• From Rain Generation to Rain Removal
• Generative Models for Security: Attacks, Defenses, and Opportunities
• Pairwise-GAN: Pose-based View Synthesis through Pair-Wise Training
• Learning Signal-Agnostic Manifolds of Neural Fields
• Improving Text to Image Generation using Mode-seeking Function
• Channel-Directed Gradients for Optimization of Convolutional Neural Networks
• WAS-VTON: Warping Architecture Search for Virtual Try-on Network
• Modeling Artistic Workflows for Image Generation and Editing
• Lifting 2D StyleGAN for 3D-Aware Face Generation
• Semi-supervised mp-MRI Data Synthesis with StitchLayer and Auxiliary Distance Maximization
• Synthesizing 3D Shapes from Silhouette Image Collections using Multi-projection Generative Adversarial Networks
• Generative Model without Prior Distribution Matching
• Defect-GAN: High-Fidelity Defect Synthesis for Automated Defect Inspection
• Encoder-Powered Generative Adversarial Networks
• AE-OT-GAN: Training GANs from data specific latent distribution
• Deep learning-based bias transfer for overcoming laboratory differences of microscopic images
• On the Exploitation of Neuroevolutionary Information: Analyzing the Past for a More Efficient Future
• DeepLandscape: Adversarial Modeling of Landscape Video
• DeepGIN: Deep Generative Inpainting Network for Extreme Image Inpainting
• TreeGAN: Incorporating Class Hierarchy into Image Generation
• Shapes and Context: In-the-Wild Image Synthesis & Manipulation
• Painting Outside as Inside: Edge Guided Image Outpainting via Bidirectional Rearrangement with Progressive Step Learning
• CariMe: Unpaired Caricature Generation with Multiple Exaggerations
• F-Drop&Match: GANs with a Dead Zone in the High-Frequency Domain
• An Empirical Study of Generative Models with Encoders
• Combining Attention with Flow for Person Image Synthesis
• Liquid Warping GAN with Attention: A Unified Framework for Human Image Synthesis
• LDC-VAE: A Latent Distribution Consistency Approach to Variational AutoEncoders
• Variational (Gradient) Estimate of the Score Function in Energy-based Latent Variable Models
• The Image Local Autoregressive Transformer
• Unsupervised Disentanglement of Linear-Encoded Facial Semantics
• Sampling Using Neural Networks for colorizing the grayscale images
• Novel View Synthesis on Unpaired Data by Conditional Deformable Variational Auto-Encoder
• HRVGAN: High Resolution Video Generation using Spatio-Temporal GAN
• S2IGAN: Speech-to-Image Generation via Adversarial Learning
• On the Anomalous Generalization of GANs
• Kernel Mean Matching for Content Addressability of GANs
• A Generative Approach Towards Improved Robotic Detection of Marine Litter
• Multi-Tailed, Multi-Headed, Spatial Dynamic Memory refined Text-to-Image Synthesis
• Comprehensive Facial Expression Synthesis using Human-Interpretable Language
• Unsupervised Image Transformation Learning via Generative Adversarial Networks
• MeronymNet: A Hierarchical Approach for Unified and Controllable Multi-Category Object Generation
• LI-Net: Large-Pose Identity-Preserving Face Reenactment Network
• AE-StyleGAN: Improved Training of Style-Based Auto-Encoders
• Anonymization of labeled TOF-MRA images for brain vessel segmentation using generative adversarial networks
• Self-Supervised Object Detection via Generative Image Synthesis
• Generation and Simulation of Yeast Microscopy Imagery with Deep Learning
• A Picture is Worth a Thousand Words: A Unified System for Diverse Captions and Rich Images Generation
• Multiple GAN Inversion for Exemplar-based Image-to-Image Translation
• Multi-attribute Pizza Generator: Cross-domain Attribute Control with Conditional StyleGAN
• An Empirical Study on GANs with Margin Cosine Loss and Relativistic Discriminator
• Spectral Tensor Train Parameterization of Deep Learning Layers
• Separating Content and Style for Unsupervised Image-to-Image Translation
• Joint Wasserstein Distribution Matching
• Unpaired Learning for High Dynamic Range Image Tone Mapping
• Impressions2Font: Generating Fonts by Specifying Impressions
• Improving Model Compatibility of Generative Adversarial Networks by Boundary Calibration
• On the Frequency Bias of Generative Models
• NeurInt : Learning to Interpolate through Neural ODEs
• Few-Shot Font Generation with Deep Metric Learning
• Localising In Complex Scenes Using Balanced Adversarial Adaptation
• Pixel-wise Dense Detector for Image Inpainting
• Neural Approximation of an Auto-Regressive Process through Confidence Guided Sampling
• Entropy-regularized optimal transport on multivariate normal and q-normal distributions
• MCMI: Multi-Cycle Image Translation with Mutual Information Constraints
• Generative Max-Mahalanobis Classifiers for Image Classification, Generation and More
• Composition and decomposition of GANs
• MTAdam: Automatic Balancing of Multiple Training Loss Terms
• Human Pose Transfer by Adaptive Hierarchical Deformation
• Virtual Conditional Generative Adversarial Networks
• Generate and Verify: Semantically Meaningful Formal Analysis of Neural Network Perception Systems
• An Empirical Study of the Effects of Sample-Mixing Methods for Efficient Training of Generative Adversarial Networks
• Enhance Convolutional Neural Networks with Noise Incentive Block
• Face-to-Music Translation Using a Distance-Preserving Generative Adversarial Network with an Auxiliary Discriminator
• ArrowGAN : Learning to Generate Videos by Learning Arrow of Time
• Efficient Semantic Image Synthesis via Class-Adaptive Normalization
• MMCGAN: Generative Adversarial Network with Explicit Manifold Prior
• Adversarial Machine Learning in Text Analysis and Generation
• Continual Learning of Generative Models with Limited Data: From Wasserstein-1 Barycenter to Adaptive Coalescence
• EBMs Trained with Maximum Likelihood are Generator Models Trained with a Self-adverserial Loss
• MUSE: Textual Attributes Guided Portrait Painting Generation
• NTIRE 2020 Challenge on Video Quality Mapping: Methods and Results
• Multi-view Alignment and Generation in CCA via Consistent Latent Encoding
• Toward a Generalization Metric for Deep Generative Models
• Unsupervised Cross-Domain Speech-to-Speech Conversion with Time-Frequency Consistency
• One Shot Audio to Animated Video Generation
• Domain Adaptation for Learning Generator from Paired Few-Shot Data
• An Edge Information and Mask Shrinking Based Image Inpainting Approach
• Synthetic Expressions are Better Than Real for Learning to Detect Facial Actions
• Host-Pathongen Co-evolution Inspired Algorithm Enables Robust GAN Training
• Hard-label Manifolds: Unexpected Advantages of Query Efficiency for Finding On-manifold Adversarial Examples
• The Implicit Metropolis-Hastings Algorithm
• APB2Face: Audio-guided face reenactment with auxiliary pose and blink signals
• Continual Density Ratio Estimation in an Online Setting
• Multichannel Generative Language Model: Learning All Possible Factorizations Within and Across Channels
• Uncertainty in Neural Processes
• Nested Scale Editing for Conditional Image Synthesis
• From Real to Synthetic and Back: Synthesizing Training Data for Multi-Person Scene Understanding
• ID-Unet: Iterative Soft and Hard Deformation for View Synthesis
• Piggyback GAN: Efficient Lifelong Learning for Image Conditioned Generation
• DeFLOCNet: Deep Image Editing via Flexible Low-level Controls
• Cascaded Refinement Network for Point Cloud Completion with Self-supervision
• LEGAN: Disentangled Manipulation of Directional Lighting and Facial Expressions by Leveraging Human Perceptual Judgements
• Investigating Conceptual Blending of a Diffusion Model for Improving Nonword-to-Image Generation
• Stay Positive: Non-Negative Image Synthesis for Augmented Reality
• Synthesizing Photorealistic Images with Deep Generative Learning
• ByeGlassesGAN: Identity Preserving Eyeglasses Removal for Face Images
• Semantic View Synthesis
• Learning Image Attacks toward Vision Guided Autonomous Vehicles
• Semantic Relation Preserving Knowledge Distillation for Image-to-Image Translation
• Head2HeadFS: Video-based Head Reenactment with Few-shot Learning
• SmartPatch: Improving Handwritten Word Imitation with Patch Discriminators
• 3D Dense Geometry-Guided Facial Expression Synthesis by Adversarial Learning
• Improving Stability of LS-GANs for Audio and Speech Signals
• CAMS: Color-Aware Multi-Style Transfer
• Bridging the Visual Gap: Wide-Range Image Blending
• Improving Relational Regularized Autoencoders with Spherical Sliced Fused Gromov Wasserstein
• Conditional Transferring Features: Scaling GANs to Thousands of Classes with 30% Less High-quality Data for Training
• Stacked Wasserstein Autoencoder
• On the Generative Utility of Cyclic Conditionals
• Semantic Example Guided Image-to-Image Translation
• Multiclass non-Adversarial Image Synthesis, with Application to Classification from Very Small Sample
• ReMix: Towards Image-to-Image Translation with Limited Data
• 3D-Aware Semantic-Guided Generative Model for Human Synthesis
• Forward Operator Estimation in Generative Models with Kernel Transfer Operators
• SpaceEdit: Learning a Unified Editing Space for Open-Domain Image Editing
• Re-designing cities with conditional adversarial networks
• Trust the Critics: Generatorless and Multipurpose WGANs with Initial Convergence Guarantees
• Learning Wildfire Model from Incomplete State Observations
• Gated SwitchGAN for multi-domain facial image translation
• Barcode Method for Generative Model Evaluation driven by Topological Data Analysis
• LinesToFacePhoto: Face Photo Generation from Lines with Conditional Self-Attention Generative Adversarial Network
• MixSyn: Learning Composition and Style for Multi-Source Image Synthesis
• Delving into Rectifiers in Style-Based Image Translation
• Generative Adversarial Networks for Astronomical Images Generation
• Invexifying Regularization of Non-Linear Least-Squares Problems
• Xp-GAN: Unsupervised Multi-object Controllable Video Generation
• MOC-GAN: Mixing Objects and Captions to Generate Realistic Images
• Translate the Facial Regions You Like Using Region-Wise Normalization
• Establishing an Evaluation Metric to Quantify Climate Change Image Realism
• Data Augmentation using Random Image Cropping for High-resolution Virtual Try-On (VITON-CROP)
• Multimodal Image Outpainting With Regularized Normalized Diversification
• Contrastive Feature Loss for Image Prediction
• ViT-Inception-GAN for Image Colourising
• Wavelets to the Rescue: Improving Sample Quality of Latent Variable Deep Generative Models
• A moment-matching metric for latent variable generative models
• Self-supervised Correlation Mining Network for Person Image Generation
• Layered Controllable Video Generation
• EdiBERT, a generative model for image editing
• Generative Adversarial Network for Probabilistic Forecast of Random Dynamical System
• Toward Learning a Unified Many-to-Many Mapping for Diverse Image Translation
• Ensembles of Generative Adversarial Networks for Disconnected Data
• Can semi-supervised learning reduce the amount of manual labelling required for effective radio galaxy morphology classification?
• Improving Generative Adversarial Networks with Local Coordinate Coding
• Smoothing the Disentangled Latent Style Space for Unsupervised Image-to-Image Translation
• Unsupervised Domain Adaptation with Variational Approximation for Cardiac Segmentation
• Momentum Contrastive Autoencoder: Using Contrastive Learning for Latent Space Distribution Matching in WAE
• On the Limitations of Multimodal VAEs
• Cascading Modular Network (CAM-Net) for Multimodal Image Synthesis
• Class Balancing GAN with a Classifier in the Loop
• Bundle Networks: Fiber Bundles, Local Trivializations, and a Generative Approach to Exploring Many-to-one Maps
• Probabilistic Autoencoder using Fisher Information
• Quantifying point cloud realism through adversarially learned latent representations
• LAViTeR: Learning Aligned Visual and Textual Representations Assisted by Image and Caption Generation
• Video-to-Video Translation for Visual Speech Synthesis
• Image-to-Image Translation of Synthetic Samples for Rare Classes
• Towards Representation Learning for Atmospheric Dynamics
• Blind Image Decomposition
• Neural Scene Decoration from a Single Photograph
• Toward Zero-Shot Unsupervised Image-to-Image Translation
• Dual Projection Generative Adversarial Networks for Conditional Image Generation
• Pixel-wise Conditioning of Generative Adversarial Networks
• Spatial Content Alignment For Pose Transfer
• Unsupervised Learning of Depth and Depth-of-Field Effect from Natural Images with Aperture Rendering Generative Adversarial Networks
• Fostering Diversity in Spatial Evolutionary Generative Adversarial Networks
• Face Sketch Synthesis via Semantic-Driven Generative Adversarial Network
• Viscos Flows: Variational Schur Conditional Sampling With Normalizing Flows
• Human Annotations Improve GAN Performances
• Implicit Greedy Rank Learning in Autoencoders via Overparameterized Linear Networks
• NP-DRAW: A Non-Parametric Structured Latent Variable Model for Image Generation
• Exploiting Relationship for Complex-scene Image Generation
• Grid Partitioned Attention: Efficient TransformerApproximation with Inductive Bias for High Resolution Detail Generation
• Generative Adversarial Learning via Kernel Density Discrimination
• Controlled AutoEncoders to Generate Faces from Voices
• Local AdaGrad-Type Algorithm for Stochastic Convex-Concave Optimization
• A Systematical Solution for Face De-identification
• Learning to See by Looking at Noise
• Synthetic Periocular Iris PAI from a Small Set of Near-Infrared-Images
• Stein Latent Optimization for Generative Adversarial Networks
• CanvasVAE: Learning to Generate Vector Graphic Documents
• UniFaceGAN: A Unified Framework for Temporally Consistent Facial Video Editing
• From Continuity to Editability: Inverting GANs with Consecutive Images
• Identifying and Exploiting Structures for Reliable Deep Learning
• A novel generative reverse net assisted evolution algorithm for expensive-computational optimizations
• Transferring Knowledge with Attention Distillation for Multi-Domain Image-to-Image Translation
• Training Deep Normalizing Flow Models in Highly Incomplete Data Scenarios with Prior Regularization
• TiVGAN: Text to Image to Video Generation with Step-by-Step Evolutionary Generator
• Zoom, Enhance! Measuring Surveillance GAN Up-sampling
• Directional GAN: A Novel Conditioning Strategy for Generative Networks
• SS-CADA: A Semi-Supervised Cross-Anatomy Domain Adaptation for Coronary Artery Segmentation
• Conditional Invertible Neural Networks for Diverse Image-to-Image Translation
• Multi-Attributed and Structured Text-to-Face Synthesis
• StackGAN: Facial Image Generation Optimizations
• Monte Carlo Simulation of SDEs using GANs
• Robust Learning in Heterogeneous Contexts
• Instance-Conditioned GAN
• FA-GAN: Feature-Aware GAN for Text to Image Synthesis
• Sparse to Dense Motion Transfer for Face Image Animation
• Backpropagating through Fréchet Inception Distance
• Generating Object Stamps
• Stereo Video Reconstruction Without Explicit Depth Maps for Endoscopic Surgery
• Strategic Prediction with Latent Aggregative Games
• Using Adaptive Gradient for Texture Learning in Single-View 3D Reconstruction
• Cross-Identity Motion Transfer for Arbitrary Objects through Pose-Attentive Video Reassembling
• Eccentric Regularization: Minimizing Hyperspherical Energy without explicit projection
• Unaligned Image-to-Image Translation by Learning to Reweight
• Adversarial Code Learning for Image Generation
• One-element Batch Training by Moving Window
• Improved Image Generation via Sparse Modeling
• Iterative Alignment Flows
• Physical Context and Timing Aware Sequence Generating GANs
• Collaging Class-specific GANs for Semantic Image Synthesis
• Bridging the Gap between Label- and Reference-based Synthesis in Multi-attribute Image-to-Image Translation
• Generative Adversarial Network: Some Analytical Perspectives
• A Wasserstein Minimum Velocity Approach to Learning Unnormalized Models
• Generating Multi-type Temporal Sequences to Mitigate Class-imbalanced Problem
• The Benefits of Pairwise Discriminators for Adversarial Training
• Learning to Inpaint by Progressively Growing the Mask Regions
• LSC-GAN: Latent Style Code Modeling for Continuous Image-to-image Translation
• Learning Realistic Human Reposing using Cyclic Self-Supervision with 3D Shape, Pose, and Appearance Consistency
• Harnessing the Conditioning Sensorium for Improved Image Translation
• Diffusion Normalizing Flow