Large Scale GAN Training for High Fidelity Natural Image Synthesis

Abstract

Despite recent progress in generative image modeling, successfully generating high-resolution, diverse samples from complex datasets such as ImageNet remains an elusive goal. To this end, we train Generative Adversarial Networks at the largest scale yet attempted, and study the instabilities specific to such scale. We find that applying orthogonal regularization to the generator renders it amenable to a simple "truncation trick," allowing fine control over the trade-off between sample fidelity and variety by reducing the variance of the Generator's input. Our modifications lead to models which set the new state of the art in class-conditional image synthesis. When trained on ImageNet at 128x128 resolution, our models (BigGANs) achieve an Inception Score (IS) of 166.5 and Frechet Inception Distance (FID) of 7.4, improving over the previous best IS of 52.52 and FID of 18.6.

Cited by in corpus (912)

• Denoising Diffusion Probabilistic Models
• On the Opportunities and Risks of Foundation Models
• Diffusion Models Beat GANs on Image Synthesis
• Generative Adversarial Network in Medical Imaging: A Review
• Score-Based Generative Modeling through Stochastic Differential Equations
• Generative Modeling by Estimating Gradients of the Data Distribution
• Skillful Precipitation Nowcasting using Deep Generative Models of Radar
• Media Forensics and DeepFakes: an overview
• Training Generative Adversarial Networks with Limited Data
• Data-Efficient Image Recognition with Contrastive Predictive Coding
• Alias-Free Generative Adversarial Networks
• Blended Diffusion for Text-driven Editing of Natural Images
• GANSpace: Discovering Interpretable GAN Controls
• Ensemble Distillation for Robust Model Fusion in Federated Learning
• Cascaded Diffusion Models for High Fidelity Image Generation
• Deep Learning for Deepfakes Creation and Detection: A Survey
• Improved Denoising Diffusion Probabilistic Models
• Freeform Diffractive Metagrating Design Based on Generative Adversarial Networks
• Large Scale Adversarial Representation Learning
• NVAE: A Deep Hierarchical Variational Autoencoder
• GRAF: Generative Radiance Fields for 3D-Aware Image Synthesis
• GAN-Leaks: A Taxonomy of Membership Inference Attacks against Generative Models
• Improved Precision and Recall Metric for Assessing Generative Models
• Massively Multilingual Neural Machine Translation in the Wild: Findings and Challenges
• Generative Adversarial Networks in Computer Vision: A Survey and Taxonomy
• 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
• Semantic Image Synthesis with Spatially-Adaptive Normalization
• Identification of Deep Network Generated Images Using Disparities in Color Components
• GANSynth: Adversarial Neural Audio Synthesis
• High-Fidelity Generative Image Compression
• Leveraging Frequency Analysis for Deep Fake Image Recognition
• Adversarial Text-to-Image Synthesis: A Review
• Unsupervised Discovery of Interpretable Directions in the GAN Latent Space
• StyleNeRF: A Style-based 3D-Aware Generator for High-resolution Image Synthesis
• Self-supervised Visual Feature Learning with Deep Neural Networks: A Survey
• Unmasking DeepFakes with simple Features
• A Large-scale Study of Representation Learning with the Visual Task Adaptation Benchmark
• Semantic Object Accuracy for Generative Text-to-Image Synthesis
• A Survey on Long-Tailed Visual Recognition
• Simulator-based training of generative models for the inverse design of metasurfaces
• Swapping Autoencoder for Deep Image Manipulation
• VideoGPT: Video Generation using VQ-VAE and Transformers
• Adversarial Video Generation on Complex Datasets
• Classification Accuracy Score for Conditional Generative Models
• Selective Synthetic Augmentation with HistoGAN for Improved Histopathology Image Classification
• An Empirical Model of Large-Batch Training
• GENESIS: Generative Scene Inference and Sampling with Object-Centric Latent Representations
• Deep Learning in Mobile and Wireless Networking: A Survey
• DiffWave: A Versatile Diffusion Model for Audio Synthesis
• Freeze the Discriminator: a Simple Baseline for Fine-Tuning GANs
• Rain rendering for evaluating and improving robustness to bad weather
• Image Augmentations for GAN Training
• Train Sparsely, Generate Densely: Memory-efficient Unsupervised Training of High-resolution Temporal GAN
• Lifelong Generative Modeling
• Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis
• BlockGAN: Learning 3D Object-aware Scene Representations from Unlabelled Images
• Review: Deep Learning in Electron Microscopy
• Jukebox: A Generative Model for Music
• Generating Diverse High-Fidelity Images with VQ-VAE-2
• Fixing Data Augmentation to Improve Adversarial Robustness
• High Fidelity Speech Synthesis with Adversarial Networks
• Data-Free Adversarial Distillation
• From Variational to Deterministic Autoencoders
• Denoising Diffusion Implicit Models
• InSituNet: Deep Image Synthesis for Parameter Space Exploration of Ensemble Simulations
• Learning Disentangled Representations in the Imaging Domain
• Differentiable Augmentation for Data-Efficient GAN Training
• StarGAN v2: Diverse Image Synthesis for Multiple Domains
• ContraGAN: Contrastive Learning for Conditional Image Generation
• High-Fidelity Image Generation With Fewer Labels
• Vector-quantized Image Modeling with Improved VQGAN
• Adversarial Attacks Against Deep Generative Models on Data: A Survey
• Image Synthesis with a Single (Robust) Classifier
• Efficient Attention: Attention with Linear Complexities
• Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data
• Structured Denoising Diffusion Models in Discrete State-Spaces
• Data synthesis and adversarial networks: A review and meta-analysis in cancer imaging
• Deep Polynomial Neural Networks
• Controlling generative models with continuous factors of variations
• Few-Shot Unsupervised Image-to-Image Translation
• ViTGAN: Training GANs with Vision Transformers
• Copyright in Generative Deep Learning
• Revisiting Training Strategies and Generalization Performance in Deep Metric Learning
• GODIVA: Generating Open-DomaIn Videos from nAtural Descriptions
• Deep Spatial Transformation for Pose-Guided Person Image Generation and Animation
• Interpreting the Latent Space of GANs for Semantic Face Editing
• CIPS-3D: A 3D-Aware Generator of GANs Based on Conditionally-Independent Pixel Synthesis
• Transflower: probabilistic autoregressive dance generation with multimodal attention
• Deep Generative Modelling: A Comparative Review of VAEs, GANs, Normalizing Flows, Energy-Based and Autoregressive Models
• Wavelet-Packets for Deepfake Image Analysis and Detection
• You Only Need Adversarial Supervision for Semantic Image Synthesis
• SESAME: Semantic Editing of Scenes by Adding, Manipulating or Erasing Objects
• GANmapper: geographical data translation
• Reconstructing Natural Scenes from fMRI Patterns using BigBiGAN
• Unrestricted Adversarial Examples via Semantic Manipulation
• Adversarial Robustness as a Prior for Learned Representations
• High-Resolution Daytime Translation Without Domain Labels
• Collaborative Learning for Faster StyleGAN Embedding
• A Good Image Generator Is What You Need for High-Resolution Video Synthesis
• Axial-DeepLab: Stand-Alone Axial-Attention for Panoptic Segmentation
• CopyCat: Many-to-Many Fine-Grained Prosody Transfer for Neural Text-to-Speech
• Stabilizing Generative Adversarial Networks: A Survey
• Conditional GAN for timeseries generation
• Maximum Entropy Generators for Energy-Based Models
• Closed-Form Factorization of Latent Semantics in GANs
• Knowledge Distillation in Iterative Generative Models for Improved Sampling Speed
• A Systematic Survey of Regularization and Normalization in GANs
• TileGAN: Synthesis of Large-Scale Non-Homogeneous Textures
• Unsupervised Object Segmentation by Redrawing
• Compressing GANs using Knowledge Distillation
• Quantum Hamiltonian-Based Models and the Variational Quantum Thermalizer Algorithm
• A Survey of Deep Learning for Scientific Discovery
• WaveFlow: A Compact Flow-based Model for Raw Audio
• Reducing Noise in GAN Training with Variance Reduced Extragradient
• Multi-scale Convolutional Neural Networks for Inverse Problems
• On Leveraging Pretrained GANs for Generation with Limited Data
• Evolving Normalization-Activation Layers
• Reliable Fidelity and Diversity Metrics for Generative Models
• LOGAN: Latent Optimisation for Generative Adversarial Networks
• Few-Shot Adaptation of Generative Adversarial Networks
• Normalization Techniques in Training DNNs: Methodology, Analysis and Application
• ImageBART: Bidirectional Context with Multinomial Diffusion for Autoregressive Image Synthesis
• Is Attention Better Than Matrix Decomposition?
• Train Large, Then Compress: Rethinking Model Size for Efficient Training and Inference of Transformers
• Evaluation Metrics for Conditional Image Generation
• Your GAN is Secretly an Energy-based Model and You Should use Discriminator Driven Latent Sampling
• LambdaNetworks: Modeling Long-Range Interactions Without Attention
• Generative Teaching Networks: Accelerating Neural Architecture Search by Learning to Generate Synthetic Training Data
• The Gaussian equivalence of generative models for learning with shallow neural networks
• Yes, we GAN: Applying Adversarial Techniques for Autonomous Driving
• Dreaming to Distill: Data-free Knowledge Transfer via DeepInversion
• Consistency Regularization for Generative Adversarial Networks
• RelGAN: Multi-Domain Image-to-Image Translation via Relative Attributes
• ATISS: Autoregressive Transformers for Indoor Scene Synthesis
• Prescribed Generative Adversarial Networks
• AutoGAN: Neural Architecture Search for Generative Adversarial Networks
• Large image datasets: A pyrrhic win for computer vision?
• DermGAN: Synthetic Generation of Clinical Skin Images with Pathology
• CNN-generated images are surprisingly easy to spot... for now
• Style and Pose Control for Image Synthesis of Humans from a Single Monocular View
• Thinking in Frequency: Face Forgery Detection by Mining Frequency-aware Clues
• Generative Models for Effective ML on Private, Decentralized Datasets
• Position, Padding and Predictions: A Deeper Look at Position Information in CNNs
• StyleSpace Analysis: Disentangled Controls for StyleGAN Image Generation
• Parallel Attention Network with Sequence Matching for Video Grounding
• Semantic Hierarchy Emerges in Deep Generative Representations for Scene Synthesis
• Creativity and Machine Learning: A Survey
• ReSSL: Relational Self-Supervised Learning with Weak Augmentation
• Getting a CLUE: A Method for Explaining Uncertainty Estimates
• A Variational Perspective on Diffusion-Based Generative Models and Score Matching
• Vector Quantized Diffusion Model for Text-to-Image Synthesis
• Convolution with even-sized kernels and symmetric padding
• UltraSR: Spatial Encoding is a Missing Key for Implicit Image Function-based Arbitrary-Scale Super-Resolution
• AGAN: Towards Automated Design of Generative Adversarial Networks
• Attributing Fake Images to GANs: Learning and Analyzing GAN Fingerprints
• Noise reduction for weak lensing mass mapping: An application of generative adversarial networks to Subaru Hyper Suprime-Cam first-year data
• On the "steerability" of generative adversarial networks
• Checkmate: Breaking the Memory Wall with Optimal Tensor Rematerialization
• SketchGraphs: A Large-Scale Dataset for Modeling Relational Geometry in Computer-Aided Design
• Adversarial Retriever-Ranker for dense text retrieval
• Inverse Graphics GAN: Learning to Generate 3D Shapes from Unstructured 2D Data
• Lower Dimensional Kernels for Video Discriminators
• VRKitchen: an Interactive 3D Virtual Environment for Task-oriented Learning
• Quaternion Generative Adversarial Networks
• High-Quality Self-Supervised Deep Image Denoising
• Boundless: Generative Adversarial Networks for Image Extension
• Cross-Modal Contrastive Learning for Text-to-Image Generation
• A Decentralized Parallel Algorithm for Training Generative Adversarial Nets
• Generative Adversarial Networks for Image and Video Synthesis: Algorithms and Applications
• Automated curricula through setter-solver interactions
• Self-Supervised Deep Learning on Point Clouds by Reconstructing Space
• End-to-End Adversarial Text-to-Speech
• Generative Adversarial Networks (GANs): An Overview of Theoretical Model, Evaluation Metrics, and Recent Developments
• Exploring Self-attention for Image Recognition
• Visual Indeterminacy in GAN Art
• A Tutorial on Deep Latent Variable Models of Natural Language
• Learning the Stein Discrepancy for Training and Evaluating Energy-Based Models without Sampling
• Spatially Controllable Image Synthesis with Internal Representation Collaging
• Data-Efficient Instance Generation from Instance Discrimination
• MelGAN-VC: Voice Conversion and Audio Style Transfer on arbitrarily long samples using Spectrograms
• A Large-Scale Study on Regularization and Normalization in GANs
• A Closer Look at the Optimization Landscapes of Generative Adversarial Networks
• On the Evaluation of Conditional GANs
• Imitation Learning via Off-Policy Distribution Matching
• Regularizing Class-wise Predictions via Self-knowledge Distillation
• PConv: Simple yet Effective Convolutional Layer for Generative Adversarial Network
• See through Gradients: Image Batch Recovery via GradInversion
• GS-WGAN: A Gradient-Sanitized Approach for Learning Differentially Private Generators
• Pedestrian Attribute Recognition: A Survey
• Decomposing 3D Scenes into Objects via Unsupervised Volume Segmentation
• Counterfactual Generative Networks
• Convolutional Generation of Textured 3D Meshes
• 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
• Nonlinear 3D Cosmic Web Simulation with Heavy-Tailed Generative Adversarial Networks
• Visual Intelligence through Human Interaction
• Deceive D: Adaptive Pseudo Augmentation for GAN Training with Limited Data
• Sketch-based Creativity Support Tools using Deep Learning
• Counterfactuals uncover the modular structure of deep generative models
• Large-Scale Generative Data-Free Distillation
• On the Benefit of Width for Neural Networks: Disappearance of Bad Basins
• The Lipschitz Constant of Self-Attention
• Text as Neural Operator: Image Manipulation by Text Instruction
• Why Spectral Normalization Stabilizes GANs: Analysis and Improvements
• Physics-informed GANs for Coastal Flood Visualization
• Global Texture Enhancement for Fake Face Detection in the Wild
• Fundus2Angio: A Conditional GAN Architecture for Generating Fluorescein Angiography Images from Retinal Fundus Photography
• NeRF-VAE: A Geometry Aware 3D Scene Generative Model
• Two Simple Ways to Learn Individual Fairness Metrics from Data
• Rethinking "Batch" in BatchNorm
• CPPN2GAN: Combining Compositional Pattern Producing Networks and GANs for Large-scale Pattern Generation
• Unsupervised Multimodal Representation Learning across Medical Images and Reports
• Domain Adaptation as a Problem of Inference on Graphical Models
• Exposing Previously Undetectable Faults in Deep Neural Networks
• Improved Consistency Regularization for GANs
• Gradient-free activation maximization for identifying effective stimuli
• Predicting Video with VQVAE
• Learning Generative Models across Incomparable Spaces
• Data-Driven Approach to Encoding and Decoding 3-D Crystal Structures
• Towards Real-World Blind Face Restoration with Generative Facial Prior
• High-resolution Deep Convolutional Generative Adversarial Networks
• Distilling and Transferring Knowledge via cGAN-generated Samples for Image Classification and Regression
• Recent Advances in Deep Learning for Object Detection
• Improving Inversion and Generation Diversity in StyleGAN using a Gaussianized Latent Space
• Hierarchical Autoregressive Image Models with Auxiliary Decoders
• Continuous User Authentication by Contactless Wireless Sensing
• Cross-domain Correspondence Learning for Exemplar-based Image Translation
• Ultra-High-Resolution Detector Simulation with Intra-Event Aware GAN and Self-Supervised Relational Reasoning
• Regularization Methods for Generative Adversarial Networks: An Overview of Recent Studies
• Opportunities and Challenges of Deep Learning Methods for Electrocardiogram Data: A Systematic Review
• Unsupervised Doodling and Painting with Improved SPIRAL
• The Bouncer Problem: Challenges to Remote Explainability
• Regularizing Deep Networks with Semantic Data Augmentation
• Ensemble flow reconstruction in the atmospheric boundary layer from spatially limited measurements through latent diffusion models
• MSG-GAN: Multi-Scale Gradients for Generative Adversarial Networks
• The Intrinsic Dimension of Images and Its Impact on Learning
• StyleGAN2 Distillation for Feed-forward Image Manipulation
• FCC-GAN: A Fully Connected and Convolutional Net Architecture for GANs
• MarioNETte: Few-shot Face Reenactment Preserving Identity of Unseen Targets
• GASNet: Weakly-supervised Framework for COVID-19 Lesion Segmentation
• MetaPix: Few-Shot Video Retargeting
• Controllable Person Image Synthesis with Attribute-Decomposed GAN
• How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Models
• GAN Prior Embedded Network for Blind Face Restoration in the Wild
• Neural Parametric Fokker-Planck Equations
• SARM: Sparse Autoregressive Model for Scalable Generation of Sparse Images in Particle Physics
• FBNetV3: Joint Architecture-Recipe Search using Predictor Pretraining
• Improving the Fairness of Deep Generative Models without Retraining
• SemanticAdv: Generating Adversarial Examples via Attribute-conditional Image Editing
• Dual Manifold Adversarial Robustness: Defense against Lp and non-Lp Adversarial Attacks
• QGAN: Quantized Generative Adversarial Networks
• Subsampling Generative Adversarial Networks: Density Ratio Estimation in Feature Space with Softplus Loss
• A unified framework for 21cm tomography sample generation and parameter inference with Progressively Growing GANs
• Global Self-Attention Networks for Image Recognition
• Smoothness and Stability in GANs
• Continual World: A Robotic Benchmark For Continual Reinforcement Learning
• Responsible Disclosure of Generative Models Using Scalable Fingerprinting
• Using Scene Graph Context to Improve Image Generation
• Finding an Unsupervised Image Segmenter in Each of Your Deep Generative Models
• Towards Better Understanding of Adaptive Gradient Algorithms in Generative Adversarial Nets
• Encoding Invariances in Deep Generative Models
• High Fidelity Video Prediction with Large Stochastic Recurrent Neural Networks
• A Spectral Energy Distance for Parallel Speech Synthesis
• Style-transfer GANs for bridging the domain gap in synthetic pose estimator training
• Structural-analogy from a Single Image Pair
• Fast and Provable ADMM for Learning with Generative Priors
• Exposing Deep-faked Videos by Anomalous Co-motion Pattern Detection
• Compatible and Diverse Fashion Image Inpainting
• Wasserstein Dependency Measure for Representation Learning
• Adversarial Lipschitz Regularization
• Nishimori meets Bethe: a spectral method for node classification in sparse weighted graphs
• Model Inversion Networks for Model-Based Optimization
• What You See is Not What the Network Infers: Detecting Adversarial Examples Based on Semantic Contradiction
• Deformable Generator Networks: Unsupervised Disentanglement of Appearance and Geometry
• Exploring the Evolution of GANs through Quality Diversity
• Generalised gravitational burst generation with Generative Adversarial Networks
• Learning Layout and Style Reconfigurable GANs for Controllable Image Synthesis
• Disentangled High Quality Salient Object Detection
• Adversarial Representation Active Learning
• Exploring DeshuffleGANs in Self-Supervised Generative Adversarial Networks
• Generative Latent Flow
• On Solving Minimax Optimization Locally: A Follow-the-Ridge Approach
• Some Theoretical Insights into Wasserstein GANs
• DatasetGAN: Efficient Labeled Data Factory with Minimal Human Effort
• A Spectral Regularizer for Unsupervised Disentanglement
• SGCN:Sparse Graph Convolution Network for Pedestrian Trajectory Prediction
• Style Generator Inversion for Image Enhancement and Animation
• Generative Adversarial Transformers
• On the distance between two neural networks and the stability of learning
• InMoDeGAN: Interpretable Motion Decomposition Generative Adversarial Network for Video Generation
• TorchGAN: A Flexible Framework for GAN Training and Evaluation
• A Generative Adversarial Network for AI-Aided Chair Design
• Object-aware Contrastive Learning for Debiased Scene Representation
• Compressing PDF sets using generative adversarial networks
• CorGAN: Correlation-Capturing Convolutional Generative Adversarial Networks for Generating Synthetic Healthcare Records
• Map completion from partial observation using the global structure of multiple environmental maps
• Highly-scalable, physics-informed GANs for learning solutions of stochastic PDEs
• O-GAN: Extremely Concise Approach for Auto-Encoding Generative Adversarial Networks
• Communication-Efficient Distributed Stochastic AUC Maximization with Deep Neural Networks
• Attention-Guided Generative Adversarial Networks for Unsupervised Image-to-Image Translation
• A Non-Parametric Test to Detect Data-Copying in Generative Models
• A case for new neural network smoothness constraints
• Disentangled Recurrent Wasserstein Autoencoder
• Stein Neural Sampler
• Few-shot Image Generation via Cross-domain Correspondence
• Latent Diffusion Models with Image-Derived Annotations for Enhanced AI-Assisted Cancer Diagnosis in Histopathology
• A Generative Model of Galactic Dust Emission Using Variational Inference
• The Six Fronts of the Generative Adversarial Networks
• Learning stochastic object models from medical imaging measurements by use of advanced ambient generative adversarial networks
• RGBD-GAN: Unsupervised 3D Representation Learning From Natural Image Datasets via RGBD Image Synthesis
• Top-k Training of GANs: Improving GAN Performance by Throwing Away Bad Samples
• Image Generation From Small Datasets via Batch Statistics Adaptation
• Fair Generative Modeling via Weak Supervision
• SARGAN: Spatial Attention-based Residuals for Facial Expression Manipulation
• Real or Not Real, that is the Question
• GENESIS-V2: Inferring Unordered Object Representations without Iterative Refinement
• Electra: Conditional Generative Model based Predicate-Aware Query Approximation
• Searching for Winograd-aware Quantized Networks
• Perceptual Generative Autoencoders
• RL-CycleGAN: Reinforcement Learning Aware Simulation-To-Real
• Local Class-Specific and Global Image-Level Generative Adversarial Networks for Semantic-Guided Scene Generation
• Bayesian Inference with Generative Adversarial Network Priors
• Self-Diagnosing GAN: Diagnosing Underrepresented Samples in Generative Adversarial Networks
• Non-Parametric Data-Driven Background Modelling using Conditional Probabilities
• Explanation by Progressive Exaggeration
• AI-assisted super-resolution cosmological simulations III: Time evolution
• From Inference to Generation: End-to-end Fully Self-supervised Generation of Human Face from Speech
• MobileStyleGAN: A Lightweight Convolutional Neural Network for High-Fidelity Image Synthesis
• UGAN: Untraceable GAN for Multi-Domain Face Translation
• Latent Translation: Crossing Modalities by Bridging Generative Models
• Study of Deep Generative Models for Inorganic Chemical Compositions
• Transformation-based Adversarial Video Prediction on Large-Scale Data
• Improving the Speed and Quality of GAN by Adversarial Training
• Effect of The Latent Structure on Clustering with GANs
• Statistical guarantees for generative models without domination
• Generating unseen complex scenes: are we there yet?
• Generative Models as Distributions of Functions
• Content and Colour Distillation for Learning Image Translations with the Spatial Profile Loss
• Image Morphing with Perceptual Constraints and STN Alignment
• Analyzing Differentiable Fuzzy Logic Operators
• Semantics Disentangling for Text-to-Image Generation
• Understanding Overparameterization in Generative Adversarial Networks
• Training Generative Adversarial Networks by Solving Ordinary Differential Equations
• FTGAN: A Fully-trained Generative Adversarial Networks for Text to Face Generation
• MichiGAN: Multi-Input-Conditioned Hair Image Generation for Portrait Editing
• Adversarial Generation of Time-Frequency Features with application in audio synthesis
• Hierarchical Patch VAE-GAN: Generating Diverse Videos from a Single Sample
• Spectral Regularization for Combating Mode Collapse in GANs
• Large-scale multilingual audio visual dubbing
• Energy-Based Imitation Learning
• Classification and reconstruction of optical quantum states with deep neural networks
• Class-Conditional VAE-GAN for Local-Ancestry Simulation
• Generative Convolution Layer for Image Generation
• Image Comes Dancing with Collaborative Parsing-Flow Video Synthesis
• cGANs with Multi-Hinge Loss
• Deep Learning based End-to-End Wireless Communication Systems with Conditional GAN as Unknown Channel
• Image Processing Using Multi-Code GAN Prior
• Compositional Transformers for Scene Generation
• Improving Automated Sonar Video Analysis to Notify About Jellyfish Blooms
• Unsupervised Image-to-Image Translation via Pre-trained StyleGAN2 Network
• Early-Phase Performance-Driven Design using Generative Models
• Regularizing Generative Adversarial Networks under Limited Data
• MISO: Mutual Information Loss with Stochastic Style Representations for Multimodal Image-to-Image Translation
• GAN "Steerability" without optimization
• LatteGAN: Visually Guided Language Attention for Multi-Turn Text-Conditioned Image Manipulation
• Bayesian Image Reconstruction using Deep Generative Models
• Disentangling Propagation and Generation for Video Prediction
• Improved Bilevel Model: Fast and Optimal Algorithm with Theoretical Guarantee
• Joint Generative and Contrastive Learning for Unsupervised Person Re-identification
• Improving GAN Training with Probability Ratio Clipping and Sample Reweighting
• Learning disconnected manifolds: a no GANs land
• Cascade EF-GAN: Progressive Facial Expression Editing with Local Focuses
• Understanding the (un)interpretability of natural image distributions using generative models
• PIRenderer: Controllable Portrait Image Generation via Semantic Neural Rendering
• Unsupervised Image Captioning
• Copula & Marginal Flows: Disentangling the Marginal from its Joint
• Further Analysis of Outlier Detection with Deep Generative Models
• Vision and Language: from Visual Perception to Content Creation
• Generative Feature Replay For Class-Incremental Learning
• Learning to Disentangle GAN Fingerprint for Fake Image Attribution
• TransGaGa: Geometry-Aware Unsupervised Image-to-Image Translation
• GraN-GAN: Piecewise Gradient Normalization for Generative Adversarial Networks
• Training Generative Adversarial Networks Via Turing Test
• A Deep Generative Model for Reordering Adjacency Matrices
• Positive-Unlabeled Reward Learning
• Investigating Under and Overfitting in Wasserstein Generative Adversarial Networks
• XingGAN for Person Image Generation
• PolyGAN: High-Order Polynomial Generators
• Fluid Flow Mass Transport for Generative Networks
• Order and Chaos: NTK views on DNN Normalization, Checkerboard and Boundary Artifacts
• Cycle In Cycle Generative Adversarial Networks for Keypoint-Guided Image Generation
• FairyTailor: A Multimodal Generative Framework for Storytelling
• DeshuffleGAN: A Self-Supervised GAN to Improve Structure Learning
• Demystifying Inter-Class Disentanglement
• MIDI-Sandwich2: RNN-based Hierarchical Multi-modal Fusion Generation VAE networks for multi-track symbolic music generation
• Adaptive Generation of Unrestricted Adversarial Inputs
• Task-Aware Feature Generation for Zero-Shot Compositional Learning
• Generative Art Using Neural Visual Grammars and Dual Encoders
• Meta-Amortized Variational Inference and Learning
• Discriminator Contrastive Divergence: Semi-Amortized Generative Modeling by Exploring Energy of the Discriminator
• Evaluating Robustness to Context-Sensitive Feature Perturbations of Different Granularities
• A Contrastive Learning Approach for Training Variational Autoencoder Priors
• Triple Generative Adversarial Networks
• NTIRE 2020 Challenge on Perceptual Extreme Super-Resolution: Methods and Results
• An Improved Self-supervised GAN via Adversarial Training
• High Fidelity Face Manipulation with Extreme Poses and Expressions
• AI Fairness via Domain Adaptation
• VAE-Info-cGAN: Generating Synthetic Images by Combining Pixel-level and Feature-level Geospatial Conditional Inputs
• Unsupervised 3D Shape Completion through GAN Inversion
• Deep Learning Techniques for Inverse Problems in Imaging
• Metropolis-Hastings view on variational inference and adversarial training
• Aesthetics of Neural Network Art
• Editing in Style: Uncovering the Local Semantics of GANs
• Wav2Pix: Speech-conditioned Face Generation using Generative Adversarial Networks
• EC-GAN: Low-Sample Classification using Semi-Supervised Algorithms and GANs
• AlphaGAN: Fully Differentiable Architecture Search for Generative Adversarial Networks
• Generating Images with Sparse Representations
• Twin Auxiliary Classifiers GAN
• Do Not Escape From the Manifold: Discovering the Local Coordinates on the Latent Space of GANs
• Continuous Conditional Generative Adversarial Networks: Novel Empirical Losses and Label Input Mechanisms
• Training GANs with Stronger Augmentations via Contrastive Discriminator
• Data Augmentation Using Adversarial Training for Construction-Equipment Classification
• Improved Natural Language Generation via Loss Truncation
• One-Shot Generative Domain Adaptation
• Conditional GANs with Auxiliary Discriminative Classifier
• Semantic Bottleneck Scene Generation
• Anycost GANs for Interactive Image Synthesis and Editing
• Invert and Defend: Model-based Approximate Inversion of Generative Adversarial Networks for Secure Inference
• Disentangling Adaptive Gradient Methods from Learning Rates
• Negative Data Augmentation
• A Multi-attribute Controllable Generative Model for Histopathology Image Synthesis
• Few-shot Video-to-Video Synthesis
• Group Equivariant Generative Adversarial Networks
• In-Domain GAN Inversion for Real Image Editing
• Learning Spatial Pyramid Attentive Pooling in Image Synthesis and Image-to-Image Translation
• RPGAN: GANs Interpretability via Random Routing
• Non-saturating GAN training as divergence minimization
• ExSinGAN: Learning an Explainable Generative Model from a Single Image
• Hybrid Encoding For Generating Large Scale Game Level Patterns With Local Variations
• Data-Efficient GAN Training Beyond (Just) Augmentations: A Lottery Ticket Perspective
• MineGAN: effective knowledge transfer from GANs to target domains with few images
• On Scalable and Efficient Computation of Large Scale Optimal Transport
• Teachers Do More Than Teach: Compressing Image-to-Image Models
• Conditional Generative Models for Counterfactual Explanations
• A Transferable Anti-Forensic Attack on Forensic CNNs Using A Generative Adversarial Network
• Self-Supervised GANs with Label Augmentation
• Toward Accurate and Realistic Virtual Try-on Through Shape Matching and Multiple Warps
• Self-supervised Semi-supervised Learning for Data Labeling and Quality Evaluation
• Label-Noise Robust Generative Adversarial Networks
• GANwriting: Content-Conditioned Generation of Styled Handwritten Word Images
• Positional Encoding as Spatial Inductive Bias in GANs
• Self Sparse Generative Adversarial Networks
• Self-Supervised GANs via Auxiliary Rotation Loss
• Learning the Precise Feature for Cluster Assignment
• Controllable Person Image Synthesis with Spatially-Adaptive Warped Normalization
• Fast video object segmentation with Spatio-Temporal GANs
• Orthogonal Convolutional Neural Networks
• Online Deep Clustering for Unsupervised Representation Learning
• Convolutional Normalization: Improving Deep Convolutional Network Robustness and Training
• Styleformer: Transformer based Generative Adversarial Networks with Style Vector
• Hijack-GAN: Unintended-Use of Pretrained, Black-Box GANs
• AniGAN: Style-Guided Generative Adversarial Networks for Unsupervised Anime Face Generation
• Understanding and Stabilizing GANs' Training Dynamics with Control Theory
• RetinaGAN: An Object-aware Approach to Sim-to-Real Transfer
• DUAL-GLOW: Conditional Flow-Based Generative Model for Modality Transfer
• VAEBM: A Symbiosis between Variational Autoencoders and Energy-based Models
• Image Synthesis From Reconfigurable Layout and Style
• Unleashing Transformers: Parallel Token Prediction with Discrete Absorbing Diffusion for Fast High-Resolution Image Generation from Vector-Quantized Codes
• Deep Adversarial Social Recommendation
• Conditional Adversarial Generative Flow for Controllable Image Synthesis
• On Relativistic $f$-Divergences
• Natural Image Matting via Guided Contextual Attention
• Unconstrained Facial Expression Transfer using Style-based Generator
• Causal Adversarial Network for Learning Conditional and Interventional Distributions
• TSIT: A Simple and Versatile Framework for Image-to-Image Translation
• ColdGANs: Taming Language GANs with Cautious Sampling Strategies
• FakePolisher: Making DeepFakes More Detection-Evasive by Shallow Reconstruction
• SegAttnGAN: Text to Image Generation with Segmentation Attention
• SketchyCOCO: Image Generation from Freehand Scene Sketches
• Horizontally Fused Training Array: An Effective Hardware Utilization Squeezer for Training Novel Deep Learning Models
• Enjoy Your Editing: Controllable GANs for Image Editing via Latent Space Navigation
• Self-labeled Conditional GANs
• Predicting Landsat Reflectance with Deep Generative Fusion
• Do 2D GANs Know 3D Shape? Unsupervised 3D shape reconstruction from 2D Image GANs
• Revitalizing CNN Attentions via Transformers in Self-Supervised Visual Representation Learning
• Denoising Diffusion Gamma Models
• CDCGen: Cross-Domain Conditional Generation via Normalizing Flows and Adversarial Training
• PERT: A Progressively Region-based Network for Scene Text Removal
• Revisiting Knowledge Distillation for Object Detection
• Decorating Your Own Bedroom: Locally Controlling Image Generation with Generative Adversarial Networks
• Towards Empirical Sandwich Bounds on the Rate-Distortion Function
• Generative Models as a Data Source for Multiview Representation Learning
• Robust Learning Meets Generative Models: Can Proxy Distributions Improve Adversarial Robustness?
• Model-Based Domain Generalization
• Decentralized Attribution of Generative Models
• Bowtie Networks: Generative Modeling for Joint Few-Shot Recognition and Novel-View Synthesis
• CPGAN: Full-Spectrum Content-Parsing Generative Adversarial Networks for Text-to-Image Synthesis
• PAGAN: Portfolio Analysis with Generative Adversarial Networks
• Make a Face: Towards Arbitrary High Fidelity Face Manipulation
• Learning Smooth Representation for Unsupervised Domain Adaptation
• Normalized Wasserstein Distance for Mixture Distributions with Applications in Adversarial Learning and Domain Adaptation
• Discriminator Feature-based Inference by Recycling the Discriminator of GANs
• Biphasic Learning of GANs for High-Resolution Image-to-Image Translation
• Loosely Conditioned Emulation of Global Climate Models With Generative Adversarial Networks
• Asymmetric GANs for Image-to-Image Translation
• CoCosNet v2: Full-Resolution Correspondence Learning for Image Translation
• MixNMatch: Multifactor Disentanglement and Encoding for Conditional Image Generation
• Generative Zero-shot Network Quantization
• Label-Noise Robust Multi-Domain Image-to-Image Translation
• REX: Revisiting Budgeted Training with an Improved Schedule
• StyleUV: Diverse and High-fidelity UV Map Generative Model
• High Resolution Face Editing with Masked GAN Latent Code Optimization
• Universality Theorems for Generative Models
• Automatic Horizontal Fusion for GPU Kernels
• Explainable Deep Learning for Uncovering Actionable Scientific Insights for Materials Discovery and Design
• Generative Modeling with Optimal Transport Maps
• Learning compositional functions via multiplicative weight updates
• Variational Inference with Continuously-Indexed Normalizing Flows
• GAN Cocktail: mixing GANs without dataset access
• Dual Contrastive Loss and Attention for GANs
• Compositional Video Synthesis with Action Graphs
• Attentive Normalization for Conditional Image Generation
• Scalable Balanced Training of Conditional Generative Adversarial Neural Networks on Image Data
• Diffusion-Weighted Magnetic Resonance Brain Images Generation with Generative Adversarial Networks and Variational Autoencoders: A Comparison Study
• Autoencoding Generative Adversarial Networks
• MaCow: Masked Convolutional Generative Flow
• Powers of layers for image-to-image translation
• PMC-GANs: Generating Multi-Scale High-Quality Pedestrian with Multimodal Cascaded GANs
• Learning by Turning: Neural Architecture Aware Optimisation
• Port-Hamiltonian Approach to Neural Network Training
• MP3net: coherent, minute-long music generation from raw audio with a simple convolutional GAN
• Launchpad: A Programming Model for Distributed Machine Learning Research
• Data-efficient visuomotor policy training using reinforcement learning and generative models
• On the Privacy Properties of GAN-generated Samples
• Visual Adversarial Imitation Learning using Variational Models
• Learning a low dimensional manifold of real cancer tissue with PathologyGAN
• A Method to Model Conditional Distributions with Normalizing Flows
• On Characterizing GAN Convergence Through Proximal Duality Gap
• An Acceleration Framework for High Resolution Image Synthesis
• SeismoGen: Seismic Waveform Synthesis Using Generative Adversarial Networks
• Bidirectional Generative Modeling Using Adversarial Gradient Estimation
• IMAGINE: Image Synthesis by Image-Guided Model Inversion
• Generative Hierarchical Features from Synthesizing Images
• Model Extraction and Defenses on Generative Adversarial Networks
• Visual Conceptual Blending with Large-scale Language and Vision Models
• Approximating Human Judgment of Generated Image Quality
• On the Pitfalls of Learning with Limited Data: A Facial Expression Recognition Case Study
• Automating Generative Deep Learning for Artistic Purposes: Challenges and Opportunities
• Dual Attention GANs for Semantic Image Synthesis
• Self-Supervised GAN Compression
• Deep Generative Learning via Schrödinger Bridge
• Measuring Fairness in Generative Models
• MoSen: Activity Modelling in Multiple-Occupancy Smart Homes
• DivCo: Diverse Conditional Image Synthesis via Contrastive Generative Adversarial Network
• Unsupervised multi-modal Styled Content Generation
• DeltaGAN: Towards Diverse Few-shot Image Generation with Sample-Specific Delta
• HumanGAN: A Generative Model of Humans Images
• Adaptive Video Highlight Detection by Learning from User History
• Kernel Stein Generative Modeling
• Efficient Synthesis of Compact Deep Neural Networks
• Generative Interventions for Causal Learning
• Semantically Multi-modal Image Synthesis
• Model-Based Robust Deep Learning: Generalizing to Natural, Out-of-Distribution Data
• Score-Guided Generative Adversarial Networks
• XRayGAN: Consistency-preserving Generation of X-ray Images from Radiology Reports
• Generalized Energy Based Models
• Deep Generative Learning via Variational Gradient Flow
• Learning Latent Space Energy-Based Prior Model
• Deep Learning Models in Software Requirements Engineering
• GIQA: Generated Image Quality Assessment
• Improving Robustness using Generated Data
• Single-shot Channel Pruning Based on Alternating Direction Method of Multipliers
• Taking Control of Intra-class Variation in Conditional GANs Under Weak Supervision
• Taming GANs with Lookahead-Minmax
• Convolutional Neural Network Interpretability with General Pattern Theory
• Effective Data Augmentation with Multi-Domain Learning GANs
• MOGAN: Morphologic-structure-aware Generative Learning from a Single Image
• Artist Style Transfer Via Quadratic Potential
• Selective Synthetic Augmentation with Quality Assurance
• Improved Transformer for High-Resolution GANs
• Estimation with Uncertainty via Conditional Generative Adversarial Networks
• PriorGAN: Real Data Prior for Generative Adversarial Nets
• Constant-Expansion Suffices for Compressed Sensing with Generative Priors
• Pre-train and Plug-in: Flexible Conditional Text Generation with Variational Auto-Encoders
• Deep Attentive Generative Adversarial Network for Photo-Realistic Image De-Quantization
• Deep Convolutional GANs for Car Image Generation
• AMP: Authentication of Media via Provenance
• Are You A Risk Taker? Adversarial Learning of Asymmetric Cross-Domain Alignment for Risk Tolerance Prediction
• Learning Neural Light Transport
• Sparsely Grouped Multi-task Generative Adversarial Networks for Facial Attribute Manipulation
• Understanding the Intrinsic Robustness of Image Distributions using Conditional Generative Models
• Training Generative Adversarial Networks in One Stage
• High-Resolution Complex Scene Synthesis with Transformers
• Two-way kernel matrix puncturing: towards resource-efficient PCA and spectral clustering
• Style Equalization: Unsupervised Learning of Controllable Generative Sequence Models
• Boosting Transferability of Targeted Adversarial Examples via Hierarchical Generative Networks
• MPG: A Multi-ingredient Pizza Image Generator with Conditional StyleGANs
• Refining Deep Generative Models via Discriminator Gradient Flow
• Unpaired Image-to-Image Translation via Latent Energy Transport
• Adaptive Weighted Discriminator for Training Generative Adversarial Networks
• Example-Guided Scene Image Synthesis using Masked Spatial-Channel Attention and Patch-Based Self-Supervision
• Conditional Denoising of Remote Sensing Imagery Using Cycle-Consistent Deep Generative Models
• Semi-supervised Learning using Adversarial Training with Good and Bad Samples
• Level generation and style enhancement -- deep learning for game development overview
• Image-to-Image Translation with Low Resolution Conditioning
• Mode matching in GANs through latent space learning and inversion
• Generative adversarial network based on chaotic time series
• BalaGAN: Image Translation Between Imbalanced Domains via Cross-Modal Transfer
• Towards Amortized Ranking-Critical Training for Collaborative Filtering
• Dodging DeepFake Detection via Implicit Spatial-Domain Notch Filtering
• Transforming the output of GANs by fine-tuning them with features from different datasets
• GAN Inversion for Out-of-Range Images with Geometric Transformations
• Heredity-aware Child Face Image Generation with Latent Space Disentanglement
• GANs with Variational Entropy Regularizers: Applications in Mitigating the Mode-Collapse Issue
• Class-Distinct and Class-Mutual Image Generation with GANs
• Enhance Images as You Like with Unpaired Learning
• Few-Shot Learning with Intra-Class Knowledge Transfer
• Continual Learning: Tackling Catastrophic Forgetting in Deep Neural Networks with Replay Processes
• Boosting Image Outpainting with Semantic Layout Prediction
• Learning to Factorize and Relight a City
• Learning Disentangled Representations with Latent Variation Predictability
• Telling Creative Stories Using Generative Visual Aids
• Auditing AI models for Verified Deployment under Semantic Specifications
• Evolving Evocative 2D Views of Generated 3D Objects
• Out-of-core Training for Extremely Large-Scale Neural Networks With Adaptive Window-Based Scheduling
• GAN Memory with No Forgetting
• Dual Contradistinctive Generative Autoencoder
• Feedback Gradient Descent: Efficient and Stable Optimization with Orthogonality for DNNs
• Augmentation-Interpolative AutoEncoders for Unsupervised Few-Shot Image Generation
• Classify and Generate: Using Classification Latent Space Representations for Image Generations
• Inclusive GAN: Improving Data and Minority Coverage in Generative Models
• World-Consistent Video-to-Video Synthesis
• Inferential Wasserstein Generative Adversarial Networks
• Lessons Learned from the Training of GANs on Artificial Datasets
• MsCGAN: Multi-scale Conditional Generative Adversarial Networks for Person Image Generation
• Ranking Neural Checkpoints
• Independent Encoder for Deep Hierarchical Unsupervised Image-to-Image Translation
• MaGNET: Uniform Sampling from Deep Generative Network Manifolds Without Retraining
• Automated Seed Quality Testing System using GAN & Active Learning
• Ensembles of GANs for synthetic training data generation
• Layer-wise Conditioning Analysis in Exploring the Learning Dynamics of DNNs
• Generative-Discriminative Complementary Learning
• Capsule GAN Using Capsule Network for Generator Architecture
• Orthogonal Wasserstein GANs
• A cost-effective method for improving and re-purposing large, pre-trained GANs by fine-tuning their class-embeddings
• crea.blender: A Neural Network-Based Image Generation Game to Assess Creativity
• Instance Selection for GANs
• Double Descent and Other Interpolation Phenomena in GANs
• Guided-GAN: Adversarial Representation Learning for Activity Recognition with Wearables
• An Empirical Study of Batch Normalization and Group Normalization in Conditional Computation
• Transfer Learning with Pre-trained Conditional Generative Models
• Collaborative Sampling in Generative Adversarial Networks
• Diverse Image Generation via Self-Conditioned GANs
• A Survey on Large-scale Machine Learning
• FIS-GAN: GAN with Flow-based Importance Sampling
• Audio Codec Enhancement with Generative Adversarial Networks
• Content-Aware GAN Compression
• Kernel-Guided Training of Implicit Generative Models with Stability Guarantees
• Unselfie: Translating Selfies to Neutral-pose Portraits in the Wild
• Dynamic Tensor Rematerialization
• On Positive-Unlabeled Classification in GAN
• Towards Lifelong Self-Supervision For Unpaired Image-to-Image Translation
• A Limited-Capacity Minimax Theorem for Non-Convex Games or: How I Learned to Stop Worrying about Mixed-Nash and Love Neural Nets
• GeomCA: Geometric Evaluation of Data Representations
• A Novel Framework for Selection of GANs for an Application
• Interpreting the Latent Space of GANs via Correlation Analysis for Controllable Concept Manipulation
• Tractable Density Estimation on Learned Manifolds with Conformal Embedding Flows
• Interpolating GANs to Scaffold Autotelic Creativity
• Deconstructing Generative Adversarial Networks
• Incremental Concept Learning via Online Generative Memory Recall
• Deep Consensus Learning
• Hard Example Generation by Texture Synthesis for Cross-domain Shape Similarity Learning
• Learning Cycle-Consistent Cooperative Networks via Alternating MCMC Teaching for Unsupervised Cross-Domain Translation
• When do GANs replicate? On the choice of dataset size
• On Predicting Generalization using GANs
• DyStyle: Dynamic Neural Network for Multi-Attribute-Conditioned Style Editing
• Improving Augmentation and Evaluation Schemes for Semantic Image Synthesis
• Omni-GAN: On the Secrets of cGANs and Beyond
• Towards transformation-resilient provenance detection of digital media
• Systematic Analysis and Removal of Circular Artifacts for StyleGAN
• Facial Emotion Recognition with Noisy Multi-task Annotations
• Blur, Noise, and Compression Robust Generative Adversarial Networks
• Robust GANs against Dishonest Adversaries
• Auto-decoding Graphs
• not-so-BigGAN: Generating High-Fidelity Images on Small Compute with Wavelet-based Super-Resolution
• Few-shot Compositional Font Generation with Dual Memory
• Assessing Deep Neural Networks as Probability Estimators
• Don't Generate Me: Training Differentially Private Generative Models with Sinkhorn Divergence
• SMYRF: Efficient Attention using Asymmetric Clustering
• Slimmable Generative Adversarial Networks
• Latent Neural Differential Equations for Video Generation
• Curriculum Learning for Deep Generative Models with Clustering
• Precision-Recall Curves Using Information Divergence Frontiers
• Exploiting GAN Internal Capacity for High-Quality Reconstruction of Natural Images
• DarkGAN: Exploiting Knowledge Distillation for Comprehensible Audio Synthesis with GANs
• Controllable Orthogonalization in Training DNNs
• GANchors: Realistic Image Perturbation Distributions for Anchors Using Generative Models
• Ensembling with Deep Generative Views
• Generative Adversarial Data Programming
• Set Distribution Networks: a Generative Model for Sets of Images
• UU-Nets Connecting Discriminator and Generator for Image to Image Translation
• Gradient penalty from a maximum margin perspective
• "Best-of-Many-Samples" Distribution Matching
• Sample weighting as an explanation for mode collapse in generative adversarial networks
• Towards Photorealistic Colorization by Imagination
• Alleviation of Gradient Exploding in GANs: Fake Can Be Real
• Auto-Embedding Generative Adversarial Networks for High Resolution Image Synthesis
• GMM-Based Generative Adversarial Encoder Learning
• TinyGAN: Distilling BigGAN for Conditional Image Generation
• Robust conditional GANs under missing or uncertain labels
• Fast Approximate Geodesics for Deep Generative Models
• Interpreting Spatially Infinite Generative Models
• OMASGAN: Out-of-Distribution Minimum Anomaly Score GAN for Sample Generation on the Boundary
• Self-supervised GAN Detector
• Low-Rank Subspaces in GANs
• Provably Convergent Algorithms for Solving Inverse Problems Using Generative Models
• Robust Training Using Natural Transformation
• Adversarial Deep Learning for Online Resource Allocation
• Semi-Anchored Multi-Step Gradient Descent Ascent Method for Structured Nonconvex-Nonconcave Composite Minimax Problems
• Weakly-supervised Generative Adversarial Networks for medical image classification
• G2R Bound: A Generalization Bound for Supervised Learning from GAN-Synthetic Data
• Generation of non-stationary stochastic fields using Generative Adversarial Networks
• CoPE: Conditional image generation using Polynomial Expansions
• Exploiting Spatial Dimensions of Latent in GAN for Real-time Image Editing
• Image Captions are Natural Prompts for Text-to-Image Models
• An Overview on Generative AI at Scale with Edge-Cloud Computing
• F-Drop&Match: GANs with a Dead Zone in the High-Frequency Domain
• Fast Computational Ghost Imaging using Unpaired Deep Learning and a Constrained Generative Adversarial Network
• Toward Accurate and Realistic Outfits Visualization with Attention to Details
• D2C: Diffusion-Denoising Models for Few-shot Conditional Generation
• Generative Modeling for Multi-task Visual Learning
• Exploiting Chain Rule and Bayes' Theorem to Compare Probability Distributions
• Structured Directional Pruning via Perturbation Orthogonal Projection
• Deeper or Wider Networks of Point Clouds with Self-attention?
• S2cGAN: Semi-Supervised Training of Conditional GANs with Fewer Labels
• Random Network Distillation as a Diversity Metric for Both Image and Text Generation
• Conditioning Trick for Training Stable GANs
• Unrealistic Feature Suppression for Generative Adversarial Networks
• Discriminative Cross-Modal Data Augmentation for Medical Imaging Applications
• Generative Models for Security: Attacks, Defenses, and Opportunities
• Combining Attention with Flow for Person Image Synthesis
• TreeGAN: Incorporating Class Hierarchy into Image Generation
• Realistic Image Synthesis with Configurable 3D Scene Layouts
• Inference-InfoGAN: Inference Independence via Embedding Orthogonal Basis Expansion
• SSN: Soft Shadow Network for Image Compositing
• EMIXER: End-to-end Multimodal X-ray Generation via Self-supervision
• On the Theoretical Equivalence of Several Trade-Off Curves Assessing Statistical Proximity
• Non-Asymptotic Error Bounds for Bidirectional GANs
• Towards a Neural Graphics Pipeline for Controllable Image Generation
• Free-Form Image Inpainting via Contrastive Attention Network
• Teaching a GAN What Not to Learn
• Tessellated Wasserstein Auto-Encoders
• One-Shot Domain Adaptation For Face Generation
• Measure-conditional Discriminator with Stationary Optimum for GANs and Statistical Distance Surrogates
• Weakly Supervised Context Encoder using DICOM metadata in Ultrasound Imaging
• Generator Surgery for Compressed Sensing
• Bayesian Imaging With Data-Driven Priors Encoded by Neural Networks: Theory, Methods, and Algorithms
• Defect-GAN: High-Fidelity Defect Synthesis for Automated Defect Inspection
• Supervised Anomaly Detection via Conditional Generative Adversarial Network and Ensemble Active Learning
• Normalizing Flows with Multi-Scale Autoregressive Priors
• Fine-grained Image-to-Image Transformation towards Visual Recognition
• YuruGAN: Yuru-Chara Mascot Generator Using Generative Adversarial Networks With Clustering Small Dataset
• MaskAAE: Latent space optimization for Adversarial Auto-Encoders
• G3AN: Disentangling Appearance and Motion for Video Generation
• DLGAN: Disentangling Label-Specific Fine-Grained Features for Image Manipulation
• Neuro-Symbolic Generative Art: A Preliminary Study
• Words as Art Materials: Generating Paintings with Sequential GANs
• Robust model training and generalisation with Studentising flows
• Bridging Explicit and Implicit Deep Generative Models via Neural Stein Estimators
• Modeling Artistic Workflows for Image Generation and Editing
• Shapes and Context: In-the-Wild Image Synthesis & Manipulation
• Generative networks as inverse problems with fractional wavelet scattering networks
• Adversarial View-Consistent Learning for Monocular Depth Estimation
• Structure-Aware Human-Action Generation
• Hard-Aware Fashion Attribute Classification
• Neural Crossbreed: Neural Based Image Metamorphosis
• Inverse mapping of face GANs
• Attribute-conditioned Layout GAN for Automatic Graphic Design
• Generative Model without Prior Distribution Matching
• On the Anomalous Generalization of GANs
• A Generative Approach Towards Improved Robotic Detection of Marine Litter
• Optimal Transport Based Generative Autoencoders
• Digital phase-only holography using deep conditional generative models
• Long-Term Human Video Generation of Multiple Futures Using Poses
• Detecting GAN generated errors
• Unsupervised Adversarial Image Inpainting
• Cellular State Transformations using Generative Adversarial Networks
• AE-OT-GAN: Training GANs from data specific latent distribution
• Support-weighted Adversarial Imitation Learning
• Perturbative GAN: GAN with Perturbation Layers
• Performing Structured Improvisations with pre-trained Deep Learning Models
• Less Memory, Faster Speed: Refining Self-Attention Module for Image Reconstruction
• Enhancing Stability in Training Conditional Generative Adversarial Networks via Selective Data Matching
• LDC-VAE: A Latent Distribution Consistency Approach to Variational AutoEncoders
• Practical Transferability Estimation for Image Classification Tasks
• Stay Positive: Non-Negative Image Synthesis for Augmented Reality
• SDA: Improving Text Generation with Self Data Augmentation
• Impression Space from Deep Template Network
• Ownership and Creativity in Generative Models
• Conditional Invertible Neural Networks for Diverse Image-to-Image Translation
• Unsupervised Image Transformation Learning via Generative Adversarial Networks
• Towards GANs' Approximation Ability
• Inducing Hierarchical Compositional Model by Sparsifying Generator Network
• Latent Compass: Creation by Navigation
• Enhanced 3DMM Attribute Control via Synthetic Dataset Creation Pipeline
• NeurInt : Learning to Interpolate through Neural ODEs
• Recent Advancements in Self-Supervised Paradigms for Visual Feature Representation
• Image Decomposition and Classification through a Generative Model
• Perceiving Music Quality with GANs
• Fine-Grained Control of Artistic Styles in Image Generation
• Improved Image Generation via Sparse Modeling
• Translate the Facial Regions You Like Using Region-Wise Normalization
• Multi-attribute Pizza Generator: Cross-domain Attribute Control with Conditional StyleGAN
• CYBORG: Blending Human Saliency Into the Loss Improves Deep Learning
• Integrating Visuospatial, Linguistic and Commonsense Structure into Story Visualization
• Channel Decomposition into Painting Actions
• GL-GAN: Adaptive Global and Local Bilevel Optimization model of Image Generation
• PixelPyramids: Exact Inference Models from Lossless Image Pyramids
• MeronymNet: A Hierarchical Approach for Unified and Controllable Multi-Category Object Generation
• 3D-Aware Semantic-Guided Generative Model for Human Synthesis
• Harnessing the Conditioning Sensorium for Improved Image Translation
• Learning from Mistakes based on Class Weighting with Application to Neural Architecture Search
• LSC-GAN: Latent Style Code Modeling for Continuous Image-to-image Translation
• Toward Quantifying Ambiguities in Artistic Images
• Collaging Class-specific GANs for Semantic Image Synthesis
• Gradient-based Hyperparameter Optimization Over Long Horizons
• Privacy-Preserving Serverless Edge Learning with Decentralized Small Data
• Improving Differentiable Architecture Search with a Generative Model
• Compact Scene Graphs for Layout Composition and Patch Retrieval
• OmiTrans: generative adversarial networks based omics-to-omics translation framework
• Per Garment Capture and Synthesis for Real-time Virtual Try-on
• Can the Transformer Be Used as a Drop-in Replacement for RNNs in Text-Generating GANs?
• Transferring Knowledge with Attention Distillation for Multi-Domain Image-to-Image Translation
• Neural Generation of Blocks for Video Coding
• Conditional Transferring Features: Scaling GANs to Thousands of Classes with 30% Less High-quality Data for Training
• Backpropagating through Fréchet Inception Distance
• Generating Relevant Counter-Examples from a Positive Unlabeled Dataset for Image Classification
• Unsupervised Disentanglement without Autoencoding: Pitfalls and Future Directions
• The power of pictures: using ML assisted image generation to engage the crowd in complex socioscientific problems
• Local AdaGrad-Type Algorithm for Stochastic Convex-Concave Optimization
• Irregular Convolutional Auto-Encoder on Point Clouds
• Synthetic Periocular Iris PAI from a Small Set of Near-Infrared-Images
• Evaluating the distribution learning capabilities of GANs
• Label-Conditioned Next-Frame Video Generation with Neural Flows
• Generative Adversarial Learning via Kernel Density Discrimination
• Expression Conditional GAN for Facial Expression-to-Expression Translation
• Explorable Tone Mapping Operators
• Semantic and Geometric Unfolding of StyleGAN Latent Space
• Grid Partitioned Attention: Efficient TransformerApproximation with Inductive Bias for High Resolution Detail Generation
• Establishing an Evaluation Metric to Quantify Climate Change Image Realism
• Towards Disentangling Latent Space for Unsupervised Semantic Face Editing
• Can adversarial training learn image captioning ?
• Cross-View Exocentric to Egocentric Video Synthesis
• Discriminator Synthesis: On reusing the other half of Generative Adversarial Networks
• Learning to Generate Novel Scene Compositions from Single Images and Videos
• MixMix: All You Need for Data-Free Compression Are Feature and Data Mixing
• Human Annotations Improve GAN Performances
• Efficient Semantic Image Synthesis via Class-Adaptive Normalization
• Reparameterized Sampling for Generative Adversarial Networks
• Unsupervised Learning of Depth and Depth-of-Field Effect from Natural Images with Aperture Rendering Generative Adversarial Networks
• TransferI2I: Transfer Learning for Image-to-Image Translation from Small Datasets
• Cascading Modular Network (CAM-Net) for Multimodal Image Synthesis
• libmolgrid: GPU Accelerated Molecular Gridding for Deep Learning Applications
• Generative Max-Mahalanobis Classifiers for Image Classification, Generation and More
• Dual Projection Generative Adversarial Networks for Conditional Image Generation
• Complex Momentum for Optimization in Games
• Vanishing Curvature and the Power of Adaptive Methods in Randomly Initialized Deep Networks
• Vehicle Image Generation Going Well with The Surroundings
• Barcode Method for Generative Model Evaluation driven by Topological Data Analysis
• Neural Scene Decoration from a Single Photograph
• Learned Interpolation for 3D Generation
• An Efficient Style Virtual Try on Network for Clothing Business Industry
• Of Moments and Matching: A Game-Theoretic Framework for Closing the Imitation Gap
• Hybrid BYOL-ViT: Efficient approach to deal with small datasets
• Adversarial Code Learning for Image Generation
• Understanding Invariance via Feedforward Inversion of Discriminatively Trained Classifiers
• Deep Automodulators
• Explicit Gradient Learning
• Generating Videos of Zero-Shot Compositions of Actions and Objects
• Subspace Capsule Network
• Mean-field Analysis of Piecewise Linear Solutions for Wide ReLU Networks
• Evo* 2020 -- Late-Breaking Abstracts Volume
• Learning To Solve Differential Equations Across Initial Conditions
• MMCGAN: Generative Adversarial Network with Explicit Manifold Prior
• Why we need an AI-resilient society- Profiling Large Language Models
• Natural Image Manipulation for Autoregressive Models Using Fisher Scores
• HRFA: High-Resolution Feature-based Attack
• Information Compensation for Deep Conditional Generative Networks
• Noise Robust Generative Adversarial Networks
• Capsule GAN for Prostate MRI Super-Resolution
• Surrogate Gradient Field for Latent Space Manipulation
• Learning Semantic-Aware Dynamics for Video Prediction
• An Empirical Study of the Effects of Sample-Mixing Methods for Efficient Training of Generative Adversarial Networks
• [RE] CNN-generated images are surprisingly easy to spot...for now
• Face-to-Music Translation Using a Distance-Preserving Generative Adversarial Network with an Auxiliary Discriminator
• Training Deep Normalizing Flow Models in Highly Incomplete Data Scenarios with Prior Regularization
• Linear Semantics in Generative Adversarial Networks
• ID-Unet: Iterative Soft and Hard Deformation for View Synthesis
• Density Fixing: Simple yet Effective Regularization Method based on the Class Prior
• BigGAN-based Bayesian reconstruction of natural images from human brain activity
• When Relation Networks meet GANs: Relation GANs with Triplet Loss
• One-shot Face Reenactment Using Appearance Adaptive Normalization
• Bridging the Gap Between $f$-GANs and Wasserstein GANs
• Semantic Relation Preserving Knowledge Distillation for Image-to-Image Translation
• ClsGAN: Selective Attribute Editing Model Based On Classification Adversarial Network
• An Investigation into the Stochasticity of Batch Whitening
• ArrowGAN : Learning to Generate Videos by Learning Arrow of Time
• FusedProp: Towards Efficient Training of Generative Adversarial Networks
• 3D Topology Transformation with Generative Adversarial Networks