Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks

Abstract

Image-to-image translation is a class of vision and graphics problems where the goal is to learn the mapping between an input image and an output image using a training set of aligned image pairs. However, for many tasks, paired training data will not be available. We present an approach for learning to translate an image from a source domain $X$ to a target domain $Y$ in the absence of paired examples. Our goal is to learn a mapping $G:X\to Y$ such that the distribution of images from $G(X)$ is indistinguishable from the distribution $Y$ using an adversarial loss. Because this mapping is highly under-constrained, we couple it with an inverse mapping $F:Y\to X$ and introduce a cycle consistency loss to push $F(G(X))\approx X$ (and vice versa). Qualitative results are presented on several tasks where paired training data does not exist, including collection style transfer, object transfiguration, season transfer, photo enhancement, etc. Quantitative comparisons against several prior methods demonstrate the superiority of our approach.

An extended version of our ICCV 2017 paper, v7 fixed the typos and updated the implementation details. Code and data: https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix

References in corpus (5)

• Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling
• Energy-based Generative Adversarial Network
• Dual Learning for Machine Translation
• Disentangling factors of variation in deep representations using adversarial training
• Learning to Generate Images of Outdoor Scenes from Attributes and Semantic Layouts

Cited by in corpus (380)

• Generative Adversarial Networks: An Overview
• Deep Visual Domain Adaptation: A Survey
• ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data
• Progressive Growing of GANs for Improved Quality, Stability, and Variation
• Style Transfer from Non-Parallel Text by Cross-Alignment
• Deep Learning Methods for Reynolds-Averaged Navier-Stokes Simulations of Airfoil Flows
• Unsupervised Image-to-Image Translation Networks
• VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learning
• Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)
• Understanding the Role of Individual Units in a Deep Neural Network
• How Generative Adversarial Networks and Their Variants Work: An Overview
• Augmented CycleGAN: Learning Many-to-Many Mappings from Unpaired Data
• Embracing Imperfect Datasets: A Review of Deep Learning Solutions for Medical Image Segmentation
• Patch-based Output Space Adversarial Learning for Joint Optic Disc and Cup Segmentation
• Deep Appearance Models for Face Rendering
• Cross-Modality Deep Feature Learning for Brain Tumor Segmentation
• Differentiable Learning of Quantum Circuit Born Machine
• Generative Adversarial Networks in Computer Vision: A Survey and Taxonomy
• Generative Adversarial Network for Medical Images (MI-GAN)
• Variational Approaches for Auto-Encoding Generative Adversarial Networks
• An empirical study on evaluation metrics of generative adversarial networks
• Adversarial Examples: Attacks and Defenses for Deep Learning
• Synthesizing Tabular Data using Generative Adversarial Networks
• Non-Stationary Texture Synthesis by Adversarial Expansion
• The Reversible Residual Network: Backpropagation Without Storing Activations
• Unsupervised Machine Translation Using Monolingual Corpora Only
• Adversarial Machine Learning in Image Classification: A Survey Towards the Defender's Perspective
• An Introduction to Image Synthesis with Generative Adversarial Nets
• Deep Learning in Photoacoustic Tomography: Current approaches and future directions
• CycleGAN, a Master of Steganography
• Self-Supervised Learning for Videos: A Survey
• Deep Tone Mapping Operator for High Dynamic Range Images
• Fighting deepfakes by detecting GAN DCT anomalies
• Fixup Initialization: Residual Learning Without Normalization
• Visual enhancement of Cone-beam CT by use of CycleGAN
• HoloStain: Holographic virtual staining of individual biological cells
• Review of data analysis in vision inspection of power lines with an in-depth discussion of deep learning technology
• Learning and Inference on Generative Adversarial Quantum Circuits
• Fashion-Gen: The Generative Fashion Dataset and Challenge
• Best of Both Worlds: Transferring Knowledge from Discriminative Learning to a Generative Visual Dialog Model
• Margin Preserving Self-paced Contrastive Learning Towards Domain Adaptation for Medical Image Segmentation
• Triangle Generative Adversarial Networks
• Data Augmentation in Emotion Classification Using Generative Adversarial Networks
• On Identifiability in Transformers
• Super-resolution reconstruction of turbulent flow at various Reynolds numbers based on generative adversarial networks
• Subsurface structure analysis using computational interpretation and learning: A visual signal processing perspective
• A-NICE-MC: Adversarial Training for MCMC
• Stochastic Variational Video Prediction
• UNIT-DDPM: UNpaired Image Translation with Denoising Diffusion Probabilistic Models
• Domain Stylization: A Strong, Simple Baseline for Synthetic to Real Image Domain Adaptation
• DNA-GAN: Learning Disentangled Representations from Multi-Attribute Images
• Pedestrian-Synthesis-GAN: Generating Pedestrian Data in Real Scene and Beyond
• MIDI-VAE: Modeling Dynamics and Instrumentation of Music with Applications to Style Transfer
• Rapid seismic domain transfer: Seismic velocity inversion and modeling using deep generative neural networks
• Domain adaptation based self-correction model for COVID-19 infection segmentation in CT images
• Unsupervised Object Segmentation by Redrawing
• Self-ensembling for visual domain adaptation
• Implicit Maximum Likelihood Estimation
• A joint model of unpaired data from scRNA-seq and spatial transcriptomics for imputing missing gene expression measurements
• Unpaired Sentiment-to-Sentiment Translation: A Cycled Reinforcement Learning Approach
• I2V-GAN: Unpaired Infrared-to-Visible Video Translation
• One-Sided Unsupervised Domain Mapping
• Pseudo-healthy synthesis with pathology disentanglement and adversarial learning
• Improving the Improved Training of Wasserstein GANs: A Consistency Term and Its Dual Effect
• Generative Modeling of Turbulence
• OmniArt: Multi-task Deep Learning for Artistic Data Analysis
• Fashion Meets Computer Vision: A Survey
• Inspect, Understand, Overcome: A Survey of Practical Methods for AI Safety
• A Closer Look at Domain Shift for Deep Learning in Histopathology
• StarGAN-VC: Non-parallel many-to-many voice conversion with star generative adversarial networks
• medigan: a Python library of pretrained generative models for medical image synthesis
• Shift-Net: Image Inpainting via Deep Feature Rearrangement
• Unsupervised Text Style Transfer using Language Models as Discriminators
• Quantum Machine Learning in Drug Discovery: Applications in Academia and Pharmaceutical Industries
• Privacy-Protective-GAN for Face De-identification
• Towards High Resolution Video Generation with Progressive Growing of Sliced Wasserstein GANs
• Unsupervised Image Super-Resolution using Cycle-in-Cycle Generative Adversarial Networks
• Generative Adversarial Self-Imitation Learning
• ST-GAN: Spatial Transformer Generative Adversarial Networks for Image Compositing
• Image Captioning with Very Scarce Supervised Data: Adversarial Semi-Supervised Learning Approach
• AGAN: Towards Automated Design of Generative Adversarial Networks
• BourGAN: Generative Networks with Metric Embeddings
• Megapixel Size Image Creation using Generative Adversarial Networks
• Found in Translation: Learning Robust Joint Representations by Cyclic Translations Between Modalities
• Adversarial Generation of Training Examples: Applications to Moving Vehicle License Plate Recognition
• Stacked Conditional Generative Adversarial Networks for Jointly Learning Shadow Detection and Shadow Removal
• Monocular 3D Object Detection with Pseudo-LiDAR Point Cloud
• Multi-Modal Imitation Learning from Unstructured Demonstrations using Generative Adversarial Nets
• Perspectives on Sim2Real Transfer for Robotics: A Summary of the R:SS 2020 Workshop
• Imitation from Observation: Learning to Imitate Behaviors from Raw Video via Context Translation
• MAGAN: Aligning Biological Manifolds
• Machine learning for complete intersection Calabi-Yau manifolds: a methodological study
• On the application of generative adversarial networks for nonlinear modal analysis
• The Information Autoencoding Family: A Lagrangian Perspective on Latent Variable Generative Models
• Low-shot Learning via Covariance-Preserving Adversarial Augmentation Networks
• Unsupervised Visual Attribute Transfer with Reconfigurable Generative Adversarial Networks
• Physics-Consistent Data-driven Waveform Inversion with Adaptive Data Augmentation
• Style Transfer in Text: Exploration and Evaluation
• On Training Sample Memorization: Lessons from Benchmarking Generative Modeling with a Large-scale Competition
• Distinguishing Natural and Computer-Generated Images using Multi-Colorspace fused EfficientNet
• Text as Neural Operator: Image Manipulation by Text Instruction
• Memorization Precedes Generation: Learning Unsupervised GANs with Memory Networks
• Image to Image Translation for Domain Adaptation
• M2M-GAN: Many-to-Many Generative Adversarial Transfer Learning for Person Re-Identification
• 3D-A-Nets: 3D Deep Dense Descriptor for Volumetric Shapes with Adversarial Networks
• Generalization in Generative Adversarial Networks: A Novel Perspective from Privacy Protection
• Unsupervised Cipher Cracking Using Discrete GANs
• JointGAN: Multi-Domain Joint Distribution Learning with Generative Adversarial Nets
• FaceShop: Deep Sketch-based Face Image Editing
• Significance-aware Information Bottleneck for Domain Adaptive Semantic Segmentation
• Generative Adversarial Networks: A Survey Towards Private and Secure Applications
• Text-to-Image-to-Text Translation using Cycle Consistent Adversarial Networks
• Cali-Sketch: Stroke Calibration and Completion for High-Quality Face Image Generation from Human-Like Sketches
• Abdominal synthetic CT reconstruction with intensity projection prior for MRI-only adaptive radiotherapy
• SARM: Sparse Autoregressive Model for Scalable Generation of Sparse Images in Particle Physics
• Mode Collapse and Regularity of Optimal Transportation Maps
• Image inpainting using frequency domain priors
• Towards Ghost-free Shadow Removal via Dual Hierarchical Aggregation Network and Shadow Matting GAN
• Pro-UIGAN: Progressive Face Hallucination from Occluded Thumbnails
• Unsupervised Depth Estimation, 3D Face Rotation and Replacement
• GANsfer Learning: Combining labelled and unlabelled data for GAN based data augmentation
• Disentangled Makeup Transfer with Generative Adversarial Network
• Causal Generative Domain Adaptation Networks
• Conditional Image-to-Image Translation
• eCommerceGAN : A Generative Adversarial Network for E-commerce
• Towards a Better Understanding and Regularization of GAN Training Dynamics
• Coherent Point Drift Networks: Unsupervised Learning of Non-Rigid Point Set Registration
• Three dimensional blind image deconvolution for fluorescence microscopy using generative adversarial networks
• CariGAN: Caricature Generation through Weakly Paired Adversarial Learning
• VIGAN: Missing View Imputation with Generative Adversarial Networks
• Medical Image Harmonization Using Deep Learning Based Canonical Mapping: Toward Robust and Generalizable Learning in Imaging
• Compressing PDF sets using generative adversarial networks
• Non-Adversarial Unsupervised Word Translation
• Ranking CGANs: Subjective Control over Semantic Image Attributes
• Rethinking Spatially-Adaptive Normalization
• Data Hiding with Deep Learning: A Survey Unifying Digital Watermarking and Steganography
• RGB-based 3D Hand Pose Estimation via Privileged Learning with Depth Images
• AlignGAN: Learning to Align Cross-Domain Images with Conditional Generative Adversarial Networks
• Common Spatial Generative Adversarial Networks based EEG Data Augmentation for Cross-Subject Brain-Computer Interface
• Constrained CycleGAN for Effective Generation of Ultrasound Sector Images of Improved Spatial Resolution
• Unsupervised Video-to-Video Translation
• Improved Part Segmentation Performance by Optimising Realism of Synthetic Images using Cycle Generative Adversarial Networks
• Latent Translation: Crossing Modalities by Bridging Generative Models
• Generative Partial Multi-View Clustering
• Learning to Generate Time-Lapse Videos Using Multi-Stage Dynamic Generative Adversarial Networks
• Asymmetric Co-Teaching for Unsupervised Cross Domain Person Re-Identification
• Improving Video Generation for Multi-functional Applications
• Multi-Domain Translation by Learning Uncoupled Autoencoders
• From source to target and back: symmetric bi-directional adaptive GAN
• Optimal Unsupervised Domain Translation
• Improving Neural Machine Translation with Conditional Sequence Generative Adversarial Nets
• Toward Characteristic-Preserving Image-based Virtual Try-On Network
• Training of deep cross-modality conversion models with a small dataset, and their application in megavoltage CT to kilovoltage CT conversion
• LRSpeech: Extremely Low-Resource Speech Synthesis and Recognition
• SIGAN: A Novel Image Generation Method for Solar Cell Defect Segmentation and Augmentation
• Variational Disentanglement for Domain Generalization
• RadialGAN: Leveraging multiple datasets to improve target-specific predictive models using Generative Adversarial Networks
• SimGAN: Hybrid Simulator Identification for Domain Adaptation via Adversarial Reinforcement Learning
• Generative Single Image Reflection Separation
• Bounded Information Rate Variational Autoencoders
• RAN4IQA: Restorative Adversarial Nets for No-Reference Image Quality Assessment
• Adversarial Discriminative Heterogeneous Face Recognition
• Adversarial Synthesis Learning Enables Segmentation Without Target Modality Ground Truth
• Differential Generative Adversarial Networks: Synthesizing Non-linear Facial Variations with Limited Number of Training Data
• Improved Neural Text Attribute Transfer with Non-parallel Data
• Unsupervised 3D Reconstruction from a Single Image via Adversarial Learning
• PhishGAN: Data Augmentation and Identification of Homoglpyh Attacks
• Temporal Interpolation via Motion Field Prediction
• Frequency-Controlled Diffusion Model for Versatile Text-Guided Image-to-Image Translation
• Sentiment Transfer using Seq2Seq Adversarial Autoencoders
• Fluid Flow Mass Transport for Generative Networks
• Deep Online Video Stabilization
• Generative Adversarial Network based on Resnet for Conditional Image Restoration
• Pixel-wise Conditioned Generative Adversarial Networks for Image Synthesis and Completion
• Deep CT to MR Synthesis using Paired and Unpaired Data
• A Hierarchical Recurrent Neural Network for Symbolic Melody Generation
• Galaxy Image Translation with Semi-supervised Noise-reconstructed Generative Adversarial Networks
• Real-time Image Enhancement for Vision-based Autonomous Underwater Vehicle Navigation in Murky Waters
• Biases in Generative Art -- A Causal Look from the Lens of Art History
• Total-Body Low-Dose CT Image Denoising using Prior Knowledge Transfer Technique with Contrastive Regularization Mechanism
• Cross-domain few-shot learning with unlabelled data
• Cycle In Cycle Generative Adversarial Networks for Keypoint-Guided Image Generation
• Multi-View Data Generation Without View Supervision
• PhenDiff: Revealing Subtle Phenotypes with Diffusion Models in Real Images
• Deep Learning Techniques for Inverse Problems in Imaging
• Sketchforme: Composing Sketched Scenes from Text Descriptions for Interactive Applications
• Understanding the Effectiveness of Lipschitz-Continuity in Generative Adversarial Nets
• Surface Defect Classification in Real-Time Using Convolutional Neural Networks
• High Quality Facial Surface and Texture Synthesis via Generative Adversarial Networks
• Mix and match networks: encoder-decoder alignment for zero-pair image translation
• One-Shot Identity-Preserving Portrait Reenactment
• Inside Out: Transforming Images of Lab-Grown Plants for Machine Learning Applications in Agriculture
• Describe What to Change: A Text-guided Unsupervised Image-to-Image Translation Approach
• Towards Cross-modality Medical Image Segmentation with Online Mutual Knowledge Distillation
• Few-shot Video-to-Video Synthesis
• Object Detection using Domain Randomization and Generative Adversarial Refinement of Synthetic Images
• Pose Guided Fashion Image Synthesis Using Deep Generative Model
• GANerated Hands for Real-time 3D Hand Tracking from Monocular RGB
• Unsupervised Denoising for Satellite Imagery using Wavelet Subband CycleGAN
• Multimodal Sensor Fusion In Single Thermal image Super-Resolution
• Learning from Multi-domain Artistic Images for Arbitrary Style Transfer
• Divide-and-Conquer Adversarial Learning for High-Resolution Image and Video Enhancement
• Unsupervised Multi-Domain Image Translation with Domain-Specific Encoders/Decoders
• Language-Driven Image Style Transfer
• MM-Hand: 3D-Aware Multi-Modal Guided Hand Generative Network for 3D Hand Pose Synthesis
• MRI to CT Translation with GANs
• Handloom Design Generation Using Generative Networks
• Recapture as You Want
• Image to Image Translation : Generating maps from satellite images
• Chinese Typeface Transformation with Hierarchical Adversarial Network
• DADA: Deep Adversarial Data Augmentation for Extremely Low Data Regime Classification
• Railway Anomaly detection model using synthetic defect images generated by CycleGAN
• Investigation of Using Disentangled and Interpretable Representations for One-shot Cross-lingual Voice Conversion
• RetinaGAN: An Object-aware Approach to Sim-to-Real Transfer
• A Transferable Anti-Forensic Attack on Forensic CNNs Using A Generative Adversarial Network
• Learning Compositional Visual Concepts with Mutual Consistency
• Deep Domain-Adversarial Image Generation for Domain Generalisation
• GestureGAN for Hand Gesture-to-Gesture Translation in the Wild
• Mocycle-GAN: Unpaired Video-to-Video Translation
• Information Theory in Density Destructors
• Can Synthetic Data Improve Object Detection Results for Remote Sensing Images?
• Dual Gaussian-based Variational Subspace Disentanglement for Visible-Infrared Person Re-Identification
• BIM Hyperreality: Data Synthesis Using BIM and Hyperrealistic Rendering for Deep Learning
• A Unified Framework for Generalizable Style Transfer: Style and Content Separation
• BRULÈ: Barycenter-Regularized Unsupervised Landmark Extraction
• Adaptive Object Detection with Dual Multi-Label Prediction
• SDIT: Scalable and Diverse Cross-domain Image Translation
• Deep learning approach in multi-scale prediction of turbulent mixing-layer
• Decoupled Learning for Conditional Adversarial Networks
• Neural Abstract Style Transfer for Chinese Traditional Painting
• Hierarchical Sequence to Sequence Voice Conversion with Limited Data
• On the Diversity of Realistic Image Synthesis
• Machine learning with limited data
• Polite Dialogue Generation Without Parallel Data
• Generating Gameplay-Relevant Art Assets with Transfer Learning
• RoboCoDraw: Robotic Avatar Drawing with GAN-based Style Transfer and Time-efficient Path Optimization
• Unpaired Photo-to-Caricature Translation on Faces in the Wild
• Judge the Judges: A Large-Scale Evaluation Study of Neural Language Models for Online Review Generation
• Modular Vehicle Control for Transferring Semantic Information Between Weather Conditions Using GANs
• Learning to Read by Spelling: Towards Unsupervised Text Recognition
• Correcting differences in multi-site neuroimaging data using Generative Adversarial Networks
• Unpaired Multi-Domain Image Generation via Regularized Conditional GANs
• C4Synth: Cross-Caption Cycle-Consistent Text-to-Image Synthesis
• Convolutional Neural Network Interpretability with General Pattern Theory
• Trick or TReAT: Thematic Reinforcement for Artistic Typography
• Autonomously and Simultaneously Refining Deep Neural Network Parameters by a Bi-Generative Adversarial Network Aided Genetic Algorithm
• Customizing an Adversarial Example Generator with Class-Conditional GANs
• PhysGAN: Generating Physical-World-Resilient Adversarial Examples for Autonomous Driving
• Learning Implicit Generative Models with Theoretical Guarantees
• r-BTN: Cross-domain Face Composite and Synthesis from Limited Facial Patches
• AttS2S-VC: Sequence-to-Sequence Voice Conversion with Attention and Context Preservation Mechanisms
• Leveraging Virtual and Real Person for Unsupervised Person Re-identification
• Distilling portable Generative Adversarial Networks for Image Translation
• Feature Affinity based Pseudo Labeling for Semi-supervised Person Re-identification
• SyncGAN: Synchronize the Latent Space of Cross-modal Generative Adversarial Networks
• Improving Bi-directional Generation between Different Modalities with Variational Autoencoders
• Pipeline Generative Adversarial Networks for Facial Images Generation with Multiple Attributes
• Deep MR to CT Synthesis using Unpaired Data
• Adversarial Pixel-Level Generation of Semantic Images
• Image-to-Image Translation with Multi-Path Consistency Regularization
• Multi-source Domain Adaptation for Visual Sentiment Classification
• Unsupervised Image Super-Resolution with an Indirect Supervised Path
• BasketballGAN: Generating Basketball Play Simulation Through Sketching
• Domain-Specific Mappings for Generative Adversarial Style Transfer
• Improving Style-Content Disentanglement in Image-to-Image Translation
• Contrastive Unpaired Translation using Focal Loss for Patch Classification
• Reliable Multi-modal Medical Image-to-image Translation Independent of Pixel-wise Aligned Data
• Resembled Generative Adversarial Networks: Two Domains with Similar Attributes
• Improved Transformer for High-Resolution GANs
• ConvNets for Counting: Object Detection of Transient Phenomena in Steelpan Drums
• GANMEX: One-vs-One Attributions Guided by GAN-based Counterfactual Explanation Baselines
• Instance Enhancement Batch Normalization: an Adaptive Regulator of Batch Noise
• Play as You Like: Timbre-enhanced Multi-modal Music Style Transfer
• Learning Personalized Representation for Inverse Problems in Medical Imaging Using Deep Neural Network
• Cross-Domain Adversarial Auto-Encoder
• Face Aging with Contextual Generative Adversarial Nets
• Towards an Understanding of Neural Networks in Natural-Image Spaces
• Disentangled Face Attribute Editing via Instance-Aware Latent Space Search
• Painting on Placement: Forecasting Routing Congestion using Conditional Generative Adversarial Nets
• DotFAN: A Domain-transferred Face Augmentation Network for Pose and Illumination Invariant Face Recognition
• Automated Testing for Deep Learning Systems with Differential Behavior Criteria
• Learning to Fuse Music Genres with Generative Adversarial Dual Learning
• Parametric Synthesis of Text on Stylized Backgrounds using PGGANs
• Learning to Forecast Videos of Human Activity with Multi-granularity Models and Adaptive Rendering
• StructEdit: Learning Structural Shape Variations
• Neural Allocentric Intuitive Physics Prediction from Real Videos
• Unsupervised Typography Transfer
• Defective samples simulation through Neural Style Transfer for automatic surface defect segment
• Real-time Deep Pose Estimation with Geodesic Loss for Image-to-Template Rigid Registration
• Lesion Conditional Image Generation for Improved Segmentation of Intracranial Hemorrhage from CT Images
• StrokeGAN: Reducing Mode Collapse in Chinese Font Generation via Stroke Encoding
• Interpreting and Disentangling Feature Components of Various Complexity from DNNs
• More Information Supervised Probabilistic Deep Face Embedding Learning
• A Generative Model for Hallucinating Diverse Versions of Super Resolution Images
• Learning Cycle-Consistent Cooperative Networks via Alternating MCMC Teaching for Unsupervised Cross-Domain Translation
• Intuitive, Interactive Beard and Hair Synthesis with Generative Models
• Dilated Temporal Relational Adversarial Network for Generic Video Summarization
• Multivariate-Information Adversarial Ensemble for Scalable Joint Distribution Matching
• Cross-dataset Person Re-Identification Using Similarity Preserved Generative Adversarial Networks
• Paired 3D Model Generation with Conditional Generative Adversarial Networks
• Convolutional Networks with MuxOut Layers as Multi-rate Systems for Image Upscaling
• Deep Learning for Localization in the Lung
• Pay attention! - Robustifying a Deep Visuomotor Policy through Task-Focused Attention
• LiveStyle -- An Application to Transfer Artistic Styles
• CMR motion artifact correction using generative adversarial nets
• TET-GAN: Text Effects Transfer via Stylization and Destylization
• Pairwise Augmented GANs with Adversarial Reconstruction Loss
• Model Watermarking for Image Processing Networks
• Joint Learning of Neural Networks via Iterative Reweighted Least Squares
• Going Beyond Saliency Maps: Training Deep Models to Interpret Deep Models
• Rhythm-Flexible Voice Conversion without Parallel Data Using Cycle-GAN over Phoneme Posteriorgram Sequences
• Reproducing AmbientGAN: Generative models from lossy measurements
• Distributionally Robust Semi-Supervised Learning for People-Centric Sensing
• Fine-grained Classification using Heterogeneous Web Data and Auxiliary Categories
• Pushing the Technical Frontier: From Overwhelmingly Large Data Sets to Machine Learning
• A Tour of Convolutional Networks Guided by Linear Interpreters
• PortraitGAN for Flexible Portrait Manipulation
• Improved Surrogates in Inertial Confinement Fusion with Manifold and Cycle Consistencies
• Fine-grained Image-to-Image Transformation towards Visual Recognition
• Denoising Large-Scale Image Captioning from Alt-text Data using Content Selection Models
• Biased Embeddings from Wild Data: Measuring, Understanding and Removing
• Multi-Modal Super Resolution for Dense Microscopic Particle Size Estimation
• Overcoming Missing and Incomplete Modalities with Generative Adversarial Networks for Building Footprint Segmentation
• Generative Creativity: Adversarial Learning for Bionic Design
• Domain Adaptation via CycleGAN for Retina Segmentation in Optical Coherence Tomography
• Exploring Biases and Prejudice of Facial Synthesis via Semantic Latent Space
• MSO: Multi-Feature Space Joint Optimization Network for RGB-Infrared Person Re-Identification
• Defo-Net: Learning Body Deformation using Generative Adversarial Networks
• Convex Smoothed Autoencoder-Optimal Transport model
• Unsupervised Representation Adversarial Learning Network: from Reconstruction to Generation
• Generative Adversarial Networks for photo to Hayao Miyazaki style cartoons
• Steganography GAN: Cracking Steganography with Cycle Generative Adversarial Networks
• Unified cross-modality feature disentangler for unsupervised multi-domain MRI abdomen organs segmentation
• J-Net: Randomly weighted U-Net for audio source separation
• Learning to Model Aspects of Hearing Perception Using Neural Loss Functions
• Unsupervised Adversarial Image Inpainting
• Taming Adversarial Domain Transfer with Structural Constraints for Image Enhancement
• Bi-Directional Generation for Unsupervised Domain Adaptation
• GANILLA: Generative Adversarial Networks for Image to Illustration Translation
• Restricting Greed in Training of Generative Adversarial Network
• Maximal Information Leakage based Privacy Preserving Data Disclosure Mechanisms
• Performing Structured Improvisations with pre-trained Deep Learning Models
• Multi-Scale Recursive and Perception-Distortion Controllable Image Super-Resolution
• Temporally Coherent Video Harmonization Using Adversarial Networks
• Improving Model Compatibility of Generative Adversarial Networks by Boundary Calibration
• Selective Sampling and Mixture Models in Generative Adversarial Networks
• Batch Face Alignment using a Low-rank GAN
• Patch Based Transformation for Minimum Variance Beamformer Image Approximation Using Delay and Sum Pipeline
• CanvasGAN: A simple baseline for text to image generation by incrementally patching a canvas
• Robust Multi-Domain Mitosis Detection
• Instance Map based Image Synthesis with a Denoising Generative Adversarial Network
• Emotion Filtering at the Edge
• Conditional Transfer with Dense Residual Attention: Synthesizing traffic signs from street-view imagery
• High-Resolution Pelvic MRI Reconstruction Using a Generative Adversarial Network with Attention and Cyclic Loss
• Controlled AutoEncoders to Generate Faces from Voices
• Informative sample generation using class aware generative adversarial networks for classification of chest Xrays
• Diversity Transfer Network for Few-Shot Learning
• From Caesar Cipher to Unsupervised Learning: A New Method for Classifier Parameter Estimation
• Deconfounding age effects with fair representation learning when assessing dementia
• Cross-View Exocentric to Egocentric Video Synthesis
• Synthesizing Photorealistic Images with Deep Generative Learning
• Fuzzy Hashing as Perturbation-Consistent Adversarial Kernel Embedding
• Unsupervised Meta-learning of Figure-Ground Segmentation via Imitating Visual Effects
• Listening while Speaking and Visualizing: Improving ASR through Multimodal Chain
• Variational learning across domains with triplet information
• Generating a Fusion Image: One's Identity and Another's Shape
• Domain Translation with Conditional GANs: from Depth to RGB Face-to-Face
• ClaRe: Practical Class Incremental Learning By Remembering Previous Class Representations
• Generative Adversarial Networks for Video-to-Video Domain Adaptation
• Multimodal Generalized Zero Shot Learning for Gleason Grading using Self-Supervised Learning
• Quality-aware Unpaired Image-to-Image Translation
• GTAE: Graph-Transformer based Auto-Encoders for Linguistic-Constrained Text Style Transfer
• Learning the Prediction Distribution for Semi-Supervised Learning with Normalising Flows
• MCMI: Multi-Cycle Image Translation with Mutual Information Constraints
• BSD-GAN: Branched Generative Adversarial Network for Scale-Disentangled Representation Learning and Image Synthesis
• Deep Boosted Regression for MR to CT Synthesis
• Image Generation and Recognition (Emotions)
• Generative Adversarial Networks with Inverse Transformation Unit
• Manual-Label Free 3D Detection via An Open-Source Simulator
• Organ At Risk Segmentation with Multiple Modality