Instance Normalization: The Missing Ingredient for Fast Stylization

Abstract

It this paper we revisit the fast stylization method introduced in Ulyanov et. al. (2016). We show how a small change in the stylization architecture results in a significant qualitative improvement in the generated images. The change is limited to swapping batch normalization with instance normalization, and to apply the latter both at training and testing times. The resulting method can be used to train high-performance architectures for real-time image generation. The code will is made available on github at https://github.com/DmitryUlyanov/texture_nets. Full paper can be found at arXiv:1701.02096.

References in corpus (1)

• Texture Networks: Feed-forward Synthesis of Textures and Stylized Images

Cited by in corpus (710)

• Deep learning for time series classification: a review
• Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
• Improved Training of Wasserstein GANs
• Domain Generalization: A Survey
• Generative Modeling by Estimating Gradients of the Data Distribution
• U-shape Transformer for Underwater Image Enhancement
• A Lip Sync Expert Is All You Need for Speech to Lip Generation In The Wild
• Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges
• DeepInf: Social Influence Prediction with Deep Learning
• fastMRI: An Open Dataset and Benchmarks for Accelerated MRI
• A DIRT-T Approach to Unsupervised Domain Adaptation
• Artificial neural networks for neuroscientists: A primer
• In defence of metric learning for speaker recognition
• GRAF: Generative Radiance Fields for 3D-Aware Image Synthesis
• Learning Image-adaptive 3D Lookup Tables for High Performance Photo Enhancement in Real-time
• nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation
• TransGAN: Two Pure Transformers Can Make One Strong GAN, and That Can Scale Up
• RANet: Ranking Attention Network for Fast Video Object Segmentation
• Semantic Image Synthesis with Spatially-Adaptive Normalization
• Visual Foresight: Model-Based Deep Reinforcement Learning for Vision-Based Robotic Control
• Efficient-CapsNet: Capsule Network with Self-Attention Routing
• How Does Batch Normalization Help Optimization?
• Fast Patch-based Style Transfer of Arbitrary Style
• High-Fidelity Generative Image Compression
• FaceShifter: Towards High Fidelity And Occlusion Aware Face Swapping
• Stochastic Adversarial Video Prediction
• HeteroFL: Computation and Communication Efficient Federated Learning for Heterogeneous Clients
• MONet: Unsupervised Scene Decomposition and Representation
• ${\mathbf{C}}^2$Former: Calibrated and Complementary Transformer for RGB-Infrared Object Detection
• Parallel-Data-Free Voice Conversion Using Cycle-Consistent Adversarial Networks
• Are Labels Required for Improving Adversarial Robustness?
• Unpaired Motion Style Transfer from Video to Animation
• Differentiable Learning-to-Normalize via Switchable Normalization
• Style-Transfer via Texture-Synthesis
• ResT: An Efficient Transformer for Visual Recognition
• Swapping Autoencoder for Deep Image Manipulation
• Adversarial Video Generation on Complex Datasets
• Half a Percent of Labels is Enough: Efficient Animal Detection in UAV Imagery using Deep CNNs and Active Learning
• PrivacyNet: Semi-Adversarial Networks for Multi-attribute Face Privacy
• Ensemble Kalman Inversion: A Derivative-Free Technique For Machine Learning Tasks
• Deep Tone Mapping Operator for High Dynamic Range Images
• Swin Transformer V2: Scaling Up Capacity and Resolution
• Optimization for deep learning: theory and algorithms
• 3D U-Net Based Brain Tumor Segmentation and Survival Days Prediction
• MP-SENet: A Speech Enhancement Model with Parallel Denoising of Magnitude and Phase Spectra
• CMGAN: Conformer-based Metric GAN for Speech Enhancement
• Batch-Instance Normalization for Adaptively Style-Invariant Neural Networks
• Micro-Batch Training with Batch-Channel Normalization and Weight Standardization
• Artistic style transfer for videos and spherical images
• Supervised Community Detection with Line Graph Neural Networks
• Universal representations:The missing link between faces, text, planktons, and cat breeds
• CMGAN: Conformer-Based Metric-GAN for Monaural Speech Enhancement
• Fixup Initialization: Residual Learning Without Normalization
• Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis
• Review: Deep Learning in Electron Microscopy
• Visual enhancement of Cone-beam CT by use of CycleGAN
• Root Mean Square Layer Normalization
• Dataset Condensation with Gradient Matching
• Neural Style Transfer: A Review
• One-Class Classification: A Survey
• InstaGAN: Instance-aware Image-to-Image Translation
• Artistic Glyph Image Synthesis via One-Stage Few-Shot Learning
• Rethinking ImageNet Pre-training
• Channel Estimation for One-Bit Multiuser Massive MIMO Using Conditional GAN
• Clova Baseline System for the VoxCeleb Speaker Recognition Challenge 2020
• StarGAN v2: Diverse Image Synthesis for Multiple Domains
• Discovering state-parameter mappings in subsurface models using generative adversarial networks
• Cross-regional oil palm tree counting and detection via multi-level attention domain adaptation network
• Evaluating Prediction-Time Batch Normalization for Robustness under Covariate Shift
• Fully Automated Multi-Organ Segmentation in Abdominal Magnetic Resonance Imaging with Deep Neural Networks
• Towards a universal neural network encoder for time series
• Joint Discriminative and Generative Learning for Person Re-identification
• 3D RoI-aware U-Net for Accurate and Efficient Colorectal Tumor Segmentation
• MaskGAN: Towards Diverse and Interactive Facial Image Manipulation
• Fully Convolutional Speech Recognition
• Deep Polynomial Neural Networks
• AdamP: Slowing Down the Slowdown for Momentum Optimizers on Scale-invariant Weights
• Deep Spatial Transformation for Pose-Guided Person Image Generation and Animation
• Nonparallel Emotional Speech Conversion
• Unsupervised Cycle-consistent Generative Adversarial Networks for Pan-sharpening
• GraphNorm: A Principled Approach to Accelerating Graph Neural Network Training
• StyleBank: An Explicit Representation for Neural Image Style Transfer
• Deblurring Face Images using Uncertainty Guided Multi-Stream Semantic Networks
• On Self Modulation for Generative Adversarial Networks
• Semantic-aware Grad-GAN for Virtual-to-Real Urban Scene Adaption
• Collaborative Learning for Faster StyleGAN Embedding
• Theoretical Analysis of Auto Rate-Tuning by Batch Normalization
• DC-cycleGAN: Bidirectional CT-to-MR Synthesis from Unpaired Data
• DeeperForensics-1.0: A Large-Scale Dataset for Real-World Face Forgery Detection
• LATUP-Net: A Lightweight 3D Attention U-Net with Parallel Convolutions for Brain Tumor Segmentation
• Fast differentiable DNA and protein sequence optimization for molecular design
• Bridging the Gap Between Computational Photography and Visual Recognition
• Batch Normalization Biases Residual Blocks Towards the Identity Function in Deep Networks
• A Systematic Survey of Regularization and Normalization in GANs
• When Age-Invariant Face Recognition Meets Face Age Synthesis: A Multi-Task Learning Framework and A New Benchmark
• PoNA: Pose-guided Non-local Attention for Human Pose Transfer
• Discovering Neural Wirings
• Dual Adaptive Representation Alignment for Cross-domain Few-shot Learning
• DPFM: Deep Partial Functional Maps
• Choose a Transformer: Fourier or Galerkin
• A Survey on Deep Domain Adaptation for LiDAR Perception
• DPODv2: Dense Correspondence-Based 6 DoF Pose Estimation
• Visual Object Networks: Image Generation with Disentangled 3D Representation
• CoTr: Efficiently Bridging CNN and Transformer for 3D Medical Image Segmentation
• Comparison of Batch Normalization and Weight Normalization Algorithms for the Large-scale Image Classification
• Learning Likelihoods with Conditional Normalizing Flows
• Generalized Wasserstein Dice Score, Distributionally Robust Deep Learning, and Ranger for brain tumor segmentation: BraTS 2020 challenge
• Evolving Normalization-Activation Layers
• ReduNet: A White-box Deep Network from the Principle of Maximizing Rate Reduction
• Learning Two-View Correspondences and Geometry Using Order-Aware Network
• Dataset Condensation with Differentiable Siamese Augmentation
• Cross-dimensional transfer learning in medical image segmentation with deep learning
• Translating and Segmenting Multimodal Medical Volumes with Cycle- and Shape-Consistency Generative Adversarial Network
• NeRS: Neural Reflectance Surfaces for Sparse-view 3D Reconstruction in the Wild
• Normalization Techniques in Training DNNs: Methodology, Analysis and Application
• Photographic Text-to-Image Synthesis with a Hierarchically-nested Adversarial Network
• Robust Point Cloud Registration Framework Based on Deep Graph Matching(TPAMI Version)
• Learning Generalisable Omni-Scale Representations for Person Re-Identification
• Recovering the Wedge Modes Lost to 21-cm Foregrounds
• Video Representation Learning by Dense Predictive Coding
• Toward Real-World Super-Resolution via Adaptive Downsampling Models
• Physical Knowledge Enhanced Deep Neural Network for Sea Surface Temperature Prediction
• Shift-Net: Image Inpainting via Deep Feature Rearrangement
• Multiway Non-rigid Point Cloud Registration via Learned Functional Map Synchronization
• AutoGAN: Neural Architecture Search for Generative Adversarial Networks
• SegTransVAE: Hybrid CNN -- Transformer with Regularization for medical image segmentation
• Exploring Fine-Grained Representation and Recomposition for Cloth-Changing Person Re-Identification
• Four Things Everyone Should Know to Improve Batch Normalization
• An Exponential Learning Rate Schedule for Deep Learning
• Inferring Semantic Layout for Hierarchical Text-to-Image Synthesis
• SRPGAN: Perceptual Generative Adversarial Network for Single Image Super Resolution
• VCGAN: Video Colorization with Hybrid Generative Adversarial Network
• Progressive Pose Attention Transfer for Person Image Generation
• SRM : A Style-based Recalibration Module for Convolutional Neural Networks
• Shift: A Zero FLOP, Zero Parameter Alternative to Spatial Convolutions
• RGB2Hands: Real-Time Tracking of 3D Hand Interactions from Monocular RGB Video
• Domain Generalization with MixStyle
• Understanding Batch Normalization
• UMMAFormer: A Universal Multimodal-adaptive Transformer Framework for Temporal Forgery Localization
• Temporal-Framing Adaptive Network for Heart Sound Segmentation without Prior Knowledge of State Duration
• ATFaceGAN: Single Face Image Restoration and Recognition from Atmospheric Turbulence
• Disentangled Representations for Short-Term and Long-Term Person Re-Identification
• Rethinking the Usage of Batch Normalization and Dropout in the Training of Deep Neural Networks
• Learning joint segmentation of tissues and brain lesions from task-specific hetero-modal domain-shifted datasets
• Multi-domain stain normalization for digital pathology: A cycle-consistent adversarial network for whole slide images
• SemiFL: Semi-Supervised Federated Learning for Unlabeled Clients with Alternate Training
• Federated Multi-organ Segmentation with Inconsistent Labels
• Kronecker Attention Networks
• Passport-aware Normalization for Deep Model Protection
• Online Normalization for Training Neural Networks
• A Comprehensive and Modularized Statistical Framework for Gradient Norm Equality in Deep Neural Networks
• Generative Adversarial Networks for Image and Video Synthesis: Algorithms and Applications
• Label-set Loss Functions for Partial Supervision: Application to Fetal Brain 3D MRI Parcellation
• Gradient Centralization: A New Optimization Technique for Deep Neural Networks
• Domain-incremental Cardiac Image Segmentation with Style-oriented Replay and Domain-sensitive Feature Whitening
• Physics-Based Generative Adversarial Models for Image Restoration and Beyond
• Tensor Normalization and Full Distribution Training
• Multi-modal Deep Guided Filtering for Comprehensible Medical Image Processing
• TaskNorm: Rethinking Batch Normalization for Meta-Learning
• QuartzNet: Deep Automatic Speech Recognition with 1D Time-Channel Separable Convolutions
• ColorMapGAN: Unsupervised Domain Adaptation for Semantic Segmentation Using Color Mapping Generative Adversarial Networks
• Self-Supervised Visual Learning by Variable Playback Speeds Prediction of a Video
• Towards Stabilizing Batch Statistics in Backward Propagation of Batch Normalization
• Video Action Understanding
• Towards contrast-agnostic soft segmentation of the spinal cord
• Dim but not entirely dark: Extracting the Galactic Center Excess' source-count distribution with neural nets
• Context-aware Synthesis for Video Frame Interpolation
• Self-Ensembling with GAN-based Data Augmentation for Domain Adaptation in Semantic Segmentation
• Focal Frequency Loss for Image Reconstruction and Synthesis
• Simple yet Effective Way for Improving the Performance of GAN
• Semi-supervised Body Parsing and Pose Estimation for Enhancing Infant General Movement Assessment
• Better Compression with Deep Pre-Editing
• Can Graph Neural Networks Count Substructures?
• Revisiting CycleGAN for semi-supervised segmentation
• Pixel-wise Regression: 3D Hand Pose Estimation via Spatial-form Representation and Differentiable Decoder
• ZM-Net: Real-time Zero-shot Image Manipulation Network
• Improving Zero-shot Voice Style Transfer via Disentangled Representation Learning
• On Feature Normalization and Data Augmentation
• HINet: Half Instance Normalization Network for Image Restoration
• IBRNet: Learning Multi-View Image-Based Rendering
• Diversified Texture Synthesis with Feed-forward Networks
• Deep Image Spatial Transformation for Person Image Generation
• Normalized and Geometry-Aware Self-Attention Network for Image Captioning
• Rethinking "Batch" in BatchNorm
• 3D Shape Reconstruction from Vision and Touch
• M2M-GAN: Many-to-Many Generative Adversarial Transfer Learning for Person Re-Identification
• 3D Dense Face Alignment via Graph Convolution Networks
• Deep Volumetric Ambient Occlusion
• Neural Networks with Recurrent Generative Feedback
• X-LXMERT: Paint, Caption and Answer Questions with Multi-Modal Transformers
• Regularization Methods for Generative Adversarial Networks: An Overview of Recent Studies
• X2CT-GAN: Reconstructing CT from Biplanar X-Rays with Generative Adversarial Networks
• A Unified Multi-Phase CT Synthesis and Classification Framework for Kidney Cancer Diagnosis with Incomplete Data
• Ensemble flow reconstruction in the atmospheric boundary layer from spatially limited measurements through latent diffusion models
• Test-time Batch Statistics Calibration for Covariate Shift
• Comparing Normalization Methods for Limited Batch Size Segmentation Neural Networks
• Sill-Net: Feature Augmentation with Separated Illumination Representation
• DCAN: Dual Channel-wise Alignment Networks for Unsupervised Scene Adaptation
• Image inpainting using frequency domain priors
• Similarity-preserving Image-image Domain Adaptation for Person Re-identification
• MeshSDF: Differentiable Iso-Surface Extraction
• Task-Relevant Adversarial Imitation Learning
• MRI Cross-Modality NeuroImage-to-NeuroImage Translation
• Style Normalization and Restitution for Generalizable Person Re-identification
• A unified framework for 21cm tomography sample generation and parameter inference with Progressively Growing GANs
• CoT: Cooperative Training for Generative Modeling of Discrete Data
• Characterizing signal propagation to close the performance gap in unnormalized ResNets
• ArtFlow: Unbiased Image Style Transfer via Reversible Neural Flows
• Disentangled Makeup Transfer with Generative Adversarial Network
• DiNTS: Differentiable Neural Network Topology Search for 3D Medical Image Segmentation
• Geometric Approaches to Increase the Expressivity of Deep Neural Networks for MR Reconstruction
• Applying Visual Domain Style Transfer and Texture Synthesis Techniques to Audio - Insights and Challenges
• BYOL works even without batch statistics
• ConvS2S-VC: Fully convolutional sequence-to-sequence voice conversion
• Transfer Learning from Synthetic to Real-Noise Denoising with Adaptive Instance Normalization
• 3D Tomographic Pattern Synthesis for Enhancing the Quantification of COVID-19
• One-shot Voice Conversion by Separating Speaker and Content Representations with Instance Normalization
• SVCNet: Scribble-based Video Colorization Network with Temporal Aggregation
• 3D Photography using Context-aware Layered Depth Inpainting
• Location-aware Adaptive Normalization: A Deep Learning Approach For Wildfire Danger Forecasting
• Neural Collapse Under MSE Loss: Proximity to and Dynamics on the Central Path
• Augmented Hard Example Mining for Generalizable Person Re-Identification
• Channel Normalization in Convolutional Neural Network avoids Vanishing Gradients
• Dynamic Instance Normalization for Arbitrary Style Transfer
• Towards an Adversarially Robust Normalization Approach
• RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching
• Theoretical analysis and experimental validation of volume bias of soft Dice optimized segmentation maps in the context of inherent uncertainty
• Music Source Separation Based on a Lightweight Deep Learning Framework (DTTNET: DUAL-PATH TFC-TDF UNET)
• ResNet or DenseNet? Introducing Dense Shortcuts to ResNet
• C2FNAS: Coarse-to-Fine Neural Architecture Search for 3D Medical Image Segmentation
• Domain specific cues improve robustness of deep learning based segmentation of ct volumes
• Regional Homogeneity: Towards Learning Transferable Universal Adversarial Perturbations Against Defenses
• Geometry-Consistent Generative Adversarial Networks for One-Sided Unsupervised Domain Mapping
• IGUANe: a 3D generalizable CycleGAN for multicenter harmonization of brain MR images
• Evolutionary Neural Architecture Search for Retinal Vessel Segmentation
• Application of Ghost-DeblurGAN to Fiducial Marker Detection
• StarGAN-VC2: Rethinking Conditional Methods for StarGAN-Based Voice Conversion
• DeepOIS: Gyroscope-Guided Deep Optical Image Stabilizer Compensation
• The Effectiveness of Instance Normalization: a Strong Baseline for Single Image Dehazing
• Dual Residual Networks Leveraging the Potential of Paired Operations for Image Restoration
• Realistic Endoscopic Image Generation Method Using Virtual-to-real Image-domain Translation
• VIGAN: Missing View Imputation with Generative Adversarial Networks
• Adapted Center and Scale Prediction: More Stable and More Accurate
• Depth Estimation from Single-shot Monocular Endoscope Image Using Image Domain Adaptation And Edge-Aware Depth Estimation
• PowerNorm: Rethinking Batch Normalization in Transformers
• Cross-Iteration Batch Normalization
• Frustratingly Easy Person Re-Identification: Generalizing Person Re-ID in Practice
• Interactive Sketch & Fill: Multiclass Sketch-to-Image Translation
• Optimal Transport driven CycleGAN for Unsupervised Learning in Inverse Problems
• Gesture-to-Gesture Translation in the Wild via Category-Independent Conditional Maps
• Intriguing properties of adversarial training at scale
• SSL-QALAS: Self-Supervised Learning for Rapid Multiparameter Estimation in Quantitative MRI Using 3D-QALAS
• Learning to Globally Edit Images with Textual Description
• Mapping Instructions to Actions in 3D Environments with Visual Goal Prediction
• Rethinking Spatially-Adaptive Normalization
• MixStyle Neural Networks for Domain Generalization and Adaptation
• Dual Contrastive Learning for Unsupervised Image-to-Image Translation
• Deep Learning for Breast MRI Style Transfer with Limited Training Data
• Robust Point Cloud Registration Framework Based on Deep Graph Matching
• Funnel Activation for Visual Recognition
• VQVC+: One-Shot Voice Conversion by Vector Quantization and U-Net architecture
• Neural Sign Language Translation based on Human Keypoint Estimation
• Learning to Sketch with Shortcut Cycle Consistency
• BorderDet: Border Feature for Dense Object Detection
• GENESIS-V2: Inferring Unordered Object Representations without Iterative Refinement
• Knee Injury Detection using MRI with Efficiently-Layered Network (ELNet)
• An Adversarial Learning Approach to Medical Image Synthesis for Lesion Detection
• Brain Tumor Segmentation using 3D-CNNs with Uncertainty Estimation
• Dataset Condensation with Distribution Matching
• AdaStereo: A Simple and Efficient Approach for Adaptive Stereo Matching
• MISS GAN: A Multi-IlluStrator Style Generative Adversarial Network for image to illustration translation
• AIM 2020: Scene Relighting and Illumination Estimation Challenge
• Constrained CycleGAN for Effective Generation of Ultrasound Sector Images of Improved Spatial Resolution
• Coronary Artery Semantic Labeling using Edge Attention Graph Matching Network
• Semi-Supervised StyleGAN for Disentanglement Learning
• RL-CycleGAN: Reinforcement Learning Aware Simulation-To-Real
• Lesion-Aware Cross-Phase Attention Network for Renal Tumor Subtype Classification on Multi-Phase CT Scans
• ACE: Adapting to Changing Environments for Semantic Segmentation
• Learning to Anonymize Faces for Privacy Preserving Action Detection
• Synergistic Reconstruction and Synthesis via Generative Adversarial Networks for Accelerated Multi-Contrast MRI
• Axial multi-layer perceptron architecture for automatic segmentation of choroid plexus in multiple sclerosis
• Neural Analysis and Synthesis: Reconstructing Speech from Self-Supervised Representations
• Multiple Domain Experts Collaborative Learning: Multi-Source Domain Generalization For Person Re-Identification
• Anisotropic Stroke Control for Multiple Artists Style Transfer
• Improving Perceptual Quality by Phone-Fortified Perceptual Loss using Wasserstein Distance for Speech Enhancement
• Real-time Universal Style Transfer on High-resolution Images via Zero-channel Pruning
• LiDARNet: A Boundary-Aware Domain Adaptation Model for Point Cloud Semantic Segmentation
• HyNet: Learning Local Descriptor with Hybrid Similarity Measure and Triplet Loss
• Learning Graph Normalization for Graph Neural Networks
• Audio-Driven Dubbing for User Generated Contents via Style-Aware Semi-Parametric Synthesis
• Capsule networks with non-iterative cluster routing
• A Style-Aware Content Loss for Real-time HD Style Transfer
• Geometric and Physical Quantities Improve E(3) Equivariant Message Passing
• Classification and reconstruction of optical quantum states with deep neural networks
• Fast and Full-Resolution Light Field Deblurring using a Deep Neural Network
• Treatment Learning Causal Transformer for Noisy Image Classification
• Deep learning based projection domain metal segmentation for metal artifact reduction in cone beam computed tomography
• EllSeg-Gen, towards Domain Generalization for head-mounted eyetracking
• Shared Coupling-bridge for Weakly Supervised Local Feature Learning
• On the design of convolutional neural networks for automatic detection of Alzheimer's disease
• Unpaired Multi-modal Segmentation via Knowledge Distillation
• ReCoNet: Real-time Coherent Video Style Transfer Network
• Learning spectro-temporal representations of complex sounds with parameterized neural networks
• Building Computationally Efficient and Well-Generalizing Person Re-Identification Models with Metric Learning
• Extended Batch Normalization
• Hyperbolic Deep Neural Networks: A Survey
• Generalizable Person Re-identification with Relevance-aware Mixture of Experts
• UFO-ViT: High Performance Linear Vision Transformer without Softmax
• Momentum^2 Teacher: Momentum Teacher with Momentum Statistics for Self-Supervised Learning
• Dynamic Layer Normalization for Adaptive Neural Acoustic Modeling in Speech Recognition
• Enhancing Deformable Convolution based Video Frame Interpolation with Coarse-to-fine 3D CNN
• ChildPredictor: A Child Face Prediction Framework with Disentangled Learning
• Controlling Perceptual Factors in Neural Style Transfer
• Globally Injective ReLU Networks
• Normalized Attention Without Probability Cage
• Region-aware Adaptive Instance Normalization for Image Harmonization
• Efficient Deep Neural Networks
• Multi-Task Domain Adaptation for Deep Learning of Instance Grasping from Simulation
• Vid2Game: Controllable Characters Extracted from Real-World Videos
• Deep CT to MR Synthesis using Paired and Unpaired Data
• Instrument-To-Instrument translation: Instrumental advances drive restoration of solar observation series via deep learning
• Multimodal Transfer: A Hierarchical Deep Convolutional Neural Network for Fast Artistic Style Transfer
• Drafting and Revision: Laplacian Pyramid Network for Fast High-Quality Artistic Style Transfer
• Modality-aware Mutual Learning for Multi-modal Medical Image Segmentation
• Two Heads Are Better Than One: A Two-Stage Approach for Monaural Noise Reduction in the Complex Domain
• A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets
• BayesFT: Bayesian Optimization for Fault Tolerant Neural Network Architecture
• Quantitative Evaluation of Style Transfer
• Pose Guided Human Video Generation
• ACNe: Attentive Context Normalization for Robust Permutation-Equivariant Learning
• Batch Group Normalization
• Using deep generative neural networks to account for model errors in Markov chain Monte Carlo inversion
• Pixel-wise Conditioned Generative Adversarial Networks for Image Synthesis and Completion
• A Multi-view Multi-task Learning Framework for Multi-variate Time Series Forecasting
• Mask R-CNN with Pyramid Attention Network for Scene Text Detection
• A Densely Interconnected Network for Deep Learning Accelerated MRI
• Emotional Voice Conversion With Cycle-consistent Adversarial Network
• Single Underwater Image Enhancement Using an Analysis-Synthesis Network
• Image-to-image Translation via Hierarchical Style Disentanglement
• Neural Inverse Knitting: From Images to Manufacturing Instructions
• Adversarial Feature Augmentation and Normalization for Visual Recognition
• Multi-target Voice Conversion without Parallel Data by Adversarially Learning Disentangled Audio Representations
• Demystifying Batch Normalization in ReLU Networks: Equivalent Convex Optimization Models and Implicit Regularization
• iSEGAN: Improved Speech Enhancement Generative Adversarial Networks
• Meta Batch-Instance Normalization for Generalizable Person Re-Identification
• Deep learning for time series classification
• Image segmentation via Cellular Automata
• On Graph Neural Networks versus Graph-Augmented MLPs
• Adversarially Adaptive Normalization for Single Domain Generalization
• MASA-SR: Matching Acceleration and Spatial Adaptation for Reference-Based Image Super-Resolution
• Deep Video Inpainting Detection
• Time-Domain Speech Extraction with Spatial Information and Multi Speaker Conditioning Mechanism
• Multi-Mapping Image-to-Image Translation with Central Biasing Normalization
• Speech Enhancement using Self-Adaptation and Multi-Head Self-Attention
• Making EfficientNet More Efficient: Exploring Batch-Independent Normalization, Group Convolutions and Reduced Resolution Training
• Needles in Haystacks: On Classifying Tiny Objects in Large Images
• COCO-FUNIT: Few-Shot Unsupervised Image Translation with a Content Conditioned Style Encoder
• Rethinking Normalization and Elimination Singularity in Neural Networks
• 4D Deep Learning for Multiple Sclerosis Lesion Activity Segmentation
• SSN: Learning Sparse Switchable Normalization via SparsestMax
• Normalization of Neural Networks using Analytic Variance Propagation
• ePointDA: An End-to-End Simulation-to-Real Domain Adaptation Framework for LiDAR Point Cloud Segmentation
• Benchmarking the Robustness of Instance Segmentation Models
• Exponential Moving Average Normalization for Self-supervised and Semi-supervised Learning
• MediumVC: Any-to-any voice conversion using synthetic specific-speaker speeches as intermedium features
• HAD-Net: A Hierarchical Adversarial Knowledge Distillation Network for Improved Enhanced Tumour Segmentation Without Post-Contrast Images
• Aesthetic-Driven Image Enhancement by Adversarial Learning
• Style Normalization and Restitution for Domain Generalization and Adaptation
• Extending LOUPE for K-space Under-sampling Pattern Optimization in Multi-coil MRI
• SNAC: Speaker-normalized affine coupling layer in flow-based architecture for zero-shot multi-speaker text-to-speech
• PREDATOR: Registration of 3D Point Clouds with Low Overlap
• Contrastively Smoothed Class Alignment for Unsupervised Domain Adaptation
• Unsupervised Denoising for Satellite Imagery using Wavelet Subband CycleGAN
• Positional Normalization
• Regularizing by the Variance of the Activations' Sample-Variances
• ACNN: a Full Resolution DCNN for Medical Image Segmentation
• Gated Channel Transformation for Visual Recognition
• Convolutional Normalization: Improving Deep Convolutional Network Robustness and Training
• Deep Generative Model for Image Inpainting with Local Binary Pattern Learning and Spatial Attention
• AniGAN: Style-Guided Generative Adversarial Networks for Unsupervised Anime Face Generation
• Continuous Conversion of CT Kernel using Switchable CycleGAN with AdaIN
• Order Matters: Shuffling Sequence Generation for Video Prediction
• Oversampling Adversarial Network for Class-Imbalanced Fault Diagnosis
• AdaDM: Enabling Normalization for Image Super-Resolution
• MeshWalker: Deep Mesh Understanding by Random Walks
• Be Like Water: Robustness to Extraneous Variables Via Adaptive Feature Normalization
• Batch Normalization Preconditioning for Neural Network Training
• Domain Generalization on Efficient Acoustic Scene Classification using Residual Normalization
• Fast Object Detection in Compressed Video
• Teachers Do More Than Teach: Compressing Image-to-Image Models
• Proxy-Normalizing Activations to Match Batch Normalization while Removing Batch Dependence
• Distributionally Robust Deep Learning using Hardness Weighted Sampling
• Light Stage Super-Resolution: Continuous High-Frequency Relighting
• Fast and Accurate 3D Medical Image Segmentation with Data-swapping Method
• Controllable Person Image Synthesis with Spatially-Adaptive Warped Normalization
• Deep Mining External Imperfect Data for Chest X-ray Disease Screening
• MorphGAN: One-Shot Face Synthesis GAN for Detecting Recognition Bias
• Unsupervised MR Motion Artifact Deep Learning using Outlier-Rejecting Bootstrap Aggregation
• Generative Adversarial Frontal View to Bird View Synthesis
• Attention-based multi-channel speaker verification with ad-hoc microphone arrays
• CycleGAN-VC2: Improved CycleGAN-based Non-parallel Voice Conversion
• When Age-Invariant Face Recognition Meets Face Age Synthesis: A Multi-Task Learning Framework
• Synthesizing the degree of polarization uniformity from non-polarization-sensitive optical coherence tomography signals using a neural network
• Do Normalization Layers in a Deep ConvNet Really Need to Be Distinct?
• Exploring Explicit Domain Supervision for Latent Space Disentanglement in Unpaired Image-to-Image Translation
• Countering Noisy Labels By Learning From Auxiliary Clean Labels
• Deep Dose Plugin Towards Real-time Monte Carlo Dose Calculation Through a Deep Learning based Denoising Algorithm
• Domain-invariant Stereo Matching Networks
• Few-shot Open-set Recognition by Transformation Consistency
• AdaSample: Adaptive Sampling of Hard Positives for Descriptor Learning
• Is Texture Predictive for Age and Sex in Brain MRI?
• RF-Net: An End-to-End Image Matching Network based on Receptive Field
• Realistic Large-Scale Fine-Depth Dehazing Dataset from 3D Videos
• AdaIN-Switchable CycleGAN for Efficient Unsupervised Low-Dose CT Denoising
• Unsupervised Learning for Intrinsic Image Decomposition from a Single Image
• ManiGAN: Text-Guided Image Manipulation
• Tdcgan: Temporal Dilated Convolutional Generative Adversarial Network for End-to-end Speech Enhancement
• Spherical Motion Dynamics: Learning Dynamics of Neural Network with Normalization, Weight Decay, and SGD
• Unsupervised Learning Facial Parameter Regressor for Action Unit Intensity Estimation via Differentiable Renderer
• Privacy Leakage of SIFT Features via Deep Generative Model based Image Reconstruction
• TSIT: A Simple and Versatile Framework for Image-to-Image Translation
• Bowtie Networks: Generative Modeling for Joint Few-Shot Recognition and Novel-View Synthesis
• Exploring the Representational Power of Graph Autoencoder
• TexMesh: Reconstructing Detailed Human Texture and Geometry from RGB-D Video
• RoboCoDraw: Robotic Avatar Drawing with GAN-based Style Transfer and Time-efficient Path Optimization
• Unsupervised Enhancement of Real-World Depth Images Using Tri-Cycle GAN
• Label-Noise Robust Multi-Domain Image-to-Image Translation
• Ultrafast Photorealistic Style Transfer via Neural Architecture Search
• A Survey on Understanding, Visualizations, and Explanation of Deep Neural Networks
• Channel Equilibrium Networks for Learning Deep Representation
• Object-Guided Instance Segmentation for Biological Images
• Sequential Skip Prediction with Few-shot in Streamed Music Contents
• VocGAN: A High-Fidelity Real-time Vocoder with a Hierarchically-nested Adversarial Network
• Generative Adversarial Networks for Unpaired Voice Transformation on Impaired Speech
• End-to-End Learning Local Multi-view Descriptors for 3D Point Clouds
• Scene Change Detection Using Multiscale Cascade Residual Convolutional Neural Networks
• Discriminative-Generative Representation Learning for One-Class Anomaly Detection
• RGB-D-Fusion: Image Conditioned Depth Diffusion of Humanoid Subjects
• Progressive and Aligned Pose Attention Transfer for Person Image Generation
• Bootstrap Equilibrium and Probabilistic Speaker Representation Learning for Self-supervised Speaker Verification
• Self-supervised Video Object Segmentation by Motion Grouping
• Zero-shot Imitation Learning from Demonstrations for Legged Robot Visual Navigation
• Improving generative adversarial network inversion via fine-tuning GAN encoders
• Neural Abstract Style Transfer for Chinese Traditional Painting
• End-to-End Conditional GAN-based Architectures for Image Colourisation
• Block Shuffle: A Method for High-resolution Fast Style Transfer with Limited Memory
• Long-Term Cloth-Changing Person Re-identification
• Bayesian Multi-Scale Neural Network for Crowd Counting
• Measuring the Biases and Effectiveness of Content-Style Disentanglement
• Implicit Regularization and Convergence for Weight Normalization
• CycleGAN-VC3: Examining and Improving CycleGAN-VCs for Mel-spectrogram Conversion
• Diverse Semantic Image Synthesis via Probability Distribution Modeling
• Understanding the Tradeoffs in Client-side Privacy for Downstream Speech Tasks
• Attentive Normalization for Conditional Image Generation
• Conditional Sequential Modulation for Efficient Global Image Retouching
• A New Look at Ghost Normalization
• Single-Stage 6D Object Pose Estimation
• Live Face De-Identification in Video
• Attribute2Font: Creating Fonts You Want From Attributes
• Mixture of Pre-processing Experts Model for Noise Robust Deep Learning on Resource Constrained Platforms
• SPG-VTON: Semantic Prediction Guidance for Multi-pose Virtual Try-on
• Bag of Tricks for Neural Architecture Search
• Boosting CNN beyond Label in Inverse Problems
• PGMAN: An Unsupervised Generative Multi-adversarial Network for Pan-sharpening
• Automatic Segmentation of Organs-at-Risk from Head-and-Neck CT using Separable Convolutional Neural Network with Hard-Region-Weighted Loss
• Multiclass Spinal Cord Tumor Segmentation on MRI with Deep Learning
• Invertible Residual Network with Regularization for Effective Medical Image Segmentation
• Triggering Dark Showers with Conditional Dual Auto-Encoders
• Style Transfer Applied to Face Liveness Detection with User-Centered Models
• A Simple and Robust Framework for Cross-Modality Medical Image Segmentation applied to Vision Transformers
• Neural Stereoscopic Image Style Transfer
• Universal Face Restoration With Memorized Modulation
• NU-Class Net: A Novel Approach for Video Quality Enhancement
• An Enhanced Harmonic Densely Connected Hybrid Transformer Network Architecture for Chronic Wound Segmentation Utilising Multi-Colour Space Tensor Merging
• Generating Embroidery Patterns Using Image-to-Image Translation
• Normalization in Training U-Net for 2D Biomedical Semantic Segmentation
• Modulating Image Restoration with Continual Levels via Adaptive Feature Modification Layers
• Fast Universal Style Transfer for Artistic and Photorealistic Rendering
• Neural Mesh Flow: 3D Manifold Mesh Generation via Diffeomorphic Flows
• Curriculum By Smoothing
• MOGAN: Morphologic-structure-aware Generative Learning from a Single Image
• Overcoming Catastrophic Forgetting by Neuron-level Plasticity Control
• Annotation-Free Cardiac Vessel Segmentation via Knowledge Transfer from Retinal Images
• Robust Raw Waveform Speech Recognition Using Relevance Weighted Representations
• Unsupervised Pose-Aware Part Decomposition for 3D Articulated Objects
• SPDGAN: A Generative Adversarial Network based on SPD Manifold Learning for Automatic Image Colorization
• Evaluation of self-supervised pre-training for automatic infant movement classification using wearable movement sensors
• Leveraging SO(3)-steerable convolutions for pose-robust semantic segmentation in 3D medical data
• From Shadow Generation to Shadow Removal
• Contextual Information Enhanced Convolutional Neural Networks for Retinal Vessel Segmentation in Color Fundus Images
• Towards Autonomous Driving of Personal Mobility with Small and Noisy Dataset using Tsallis-statistics-based Behavioral Cloning
• Global and Local Alignment Networks for Unpaired Image-to-Image Translation
• 3D Pose Transfer with Correspondence Learning and Mesh Refinement
• PD-GAN: Probabilistic Diverse GAN for Image Inpainting
• Effective Data Fusion with Generalized Vegetation Index: Evidence from Land Cover Segmentation in Agriculture
• Tensor Processing Primitives: A Programming Abstraction for Efficiency and Portability in Deep Learning & HPC Workloads
• Distribution Aligned Multimodal and Multi-Domain Image Stylization
• DTGAN: Dual Attention Generative Adversarial Networks for Text-to-Image Generation
• Improving the Reconstruction of Disentangled Representation Learners via Multi-Stage Modeling
• Adaptive convolutional neural networks for k-space data interpolation in fast magnetic resonance imaging
• Encoding Robustness to Image Style via Adversarial Feature Perturbations
• URIE: Universal Image Enhancement for Visual Recognition in the Wild
• Unsupervised Disentanglement GAN for Domain Adaptive Person Re-Identification
• Generalizable Model-agnostic Semantic Segmentation via Target-specific Normalization
• Augmented Cyclic Consistency Regularization for Unpaired Image-to-Image Translation
• Batch norm with entropic regularization turns deterministic autoencoders into generative models
• Optimal Quantization for Batch Normalization in Neural Network Deployments and Beyond
• Sketch-to-Art: Synthesizing Stylized Art Images From Sketches
• Walking the Tightrope: An Investigation of the Convolutional Autoencoder Bottleneck
• A Deep Factorization of Style and Structure in Fonts
• Emotion Generation and Recognition: A StarGAN Approach
• LIP: Local Importance-based Pooling
• U-Net Training with Instance-Layer Normalization
• Very Long Natural Scenery Image Prediction by Outpainting
• Deformation-aware Unpaired Image Translation for Pose Estimation on Laboratory Animals
• Speech-to-Singing Conversion in an Encoder-Decoder Framework
• GLStyleNet: Higher Quality Style Transfer Combining Global and Local Pyramid Features
• Class-Distinct and Class-Mutual Image Generation with GANs
• Sparsely Grouped Multi-task Generative Adversarial Networks for Facial Attribute Manipulation
• Optimal Textures: Fast and Robust Texture Synthesis and Style Transfer through Optimal Transport
• Defective samples simulation through Neural Style Transfer for automatic surface defect segment
• Fashion Editing with Adversarial Parsing Learning
• How Powerful Are Randomly Initialized Pointcloud Set Functions?
• Meta-Learning Bidirectional Update Rules
• On the Periodic Behavior of Neural Network Training with Batch Normalization and Weight Decay
• SSAH: Semi-supervised Adversarial Deep Hashing with Self-paced Hard Sample Generation
• MASS: Multi-task Anthropomorphic Speech Synthesis Framework
• Linear Context Transform Block
• Approximated Orthonormal Normalisation in Training Neural Networks
• CycleQSM: Unsupervised QSM Deep Learning using Physics-Informed CycleGAN
• Towards transformation-resilient provenance detection of digital media
• Scyclone: High-Quality and Parallel-Data-Free Voice Conversion Using Spectrogram and Cycle-Consistent Adversarial Networks
• Unsupervised Multi-Domain Multimodal Image-to-Image Translation with Explicit Domain-Constrained Disentanglement
• Deep Spectral Convolution Network for HyperSpectral Unmixing
• Layer-wise Conditioning Analysis in Exploring the Learning Dynamics of DNNs
• Continuous Face Aging via Self-estimated Residual Age Embedding
• LCS: Learning Compressible Subspaces for Adaptive Network Compression at Inference Time
• Speech-to-Singing Conversion based on Boundary Equilibrium GAN
• SAFIN: Arbitrary Style Transfer With Self-Attentive Factorized Instance Normalization
• An Empirical Study of Batch Normalization and Group Normalization in Conditional Computation
• Meta-CoTGAN: A Meta Cooperative Training Paradigm for Improving Adversarial Text Generation
• Unsupervised Acoustic Unit Representation Learning for Voice Conversion using WaveNet Auto-encoders
• TET-GAN: Text Effects Transfer via Stylization and Destylization
• Is Batch Norm unique? An empirical investigation and prescription to emulate the best properties of common normalizers without batch dependence
• Learning To Pay Attention To Mistakes
• SLGAN: Style- and Latent-guided Generative Adversarial Network for Desirable Makeup Transfer and Removal
• FastSVC: Fast Cross-Domain Singing Voice Conversion with Feature-wise Linear Modulation
• 3DSNet: Unsupervised Shape-to-Shape 3D Style Transfer
• Disentangling 3D Prototypical Networks For Few-Shot Concept Learning
• Deep High-Resolution Network for Low Dose X-ray CT Denoising
• AutoML Segmentation for 3D Medical Image Data: Contribution to the MSD Challenge 2018
• CSLP-AE: A Contrastive Split-Latent Permutation Autoencoder Framework for Zero-Shot Electroencephalography Signal Conversion
• Many-to-Many Voice Conversion with Out-of-Dataset Speaker Support
• UCLID-Net: Single View Reconstruction in Object Space
• Deep Feature Response Discriminative Calibration
• MetaPerturb: Transferable Regularizer for Heterogeneous Tasks and Architectures
• Style is a Distribution of Features
• Rain Removal and Illumination Enhancement Done in One Go
• Enhancing Content Preservation in Text Style Transfer Using Reverse Attention and Conditional Layer Normalization
• Smoother Network Tuning and Interpolation for Continuous-level Image Processing
• Two Deep Learning Approaches for Automated Segmentation of Left Ventricle in Cine Cardiac MRI
• Initializing ReLU networks in an expressive subspace of weights
• Show, Attend and Translate: Unpaired Multi-Domain Image-to-Image Translation with Visual Attention
• Systematic Analysis and Removal of Circular Artifacts for StyleGAN
• Brain Age Estimation Using LSTM on Children's Brain MRI
• Deep Curiosity Loops in Social Environments
• Unsupervised Domain Generalization for Person Re-identification: A Domain-specific Adaptive Framework
• Neuron with Steady Response Leads to Better Generalization
• Towards Learning Universal Audio Representations
• ET-GAN: Cross-Language Emotion Transfer Based on Cycle-Consistent Generative Adversarial Networks
• CropDefender: deep watermark which is more convenient to train and more robust against cropping
• Interpreting Spatially Infinite Generative Models
• Flexible Image Denoising with Multi-layer Conditional Feature Modulation
• Photometric Transformer Networks and Label Adjustment for Breast Density Prediction
• Switchable Deep Beamformer
• Superpixel Segmentation via Convolutional Neural Networks with Regularized Information Maximization
• Pairwise-GAN: Pose-based View Synthesis through Pair-Wise Training
• MimicNorm: Weight Mean and Last BN Layer Mimic the Dynamic of Batch Normalization
• GAZEV: GAN-Based Zero-Shot Voice Conversion over Non-parallel Speech Corpus
• Test-Time Adaptation for Out-of-distributed Image Inpainting
• Deep Unitary Convolutional Neural Networks
• Finet: Using Fine-grained Batch Normalization to Train Light-weight Neural Networks
• Domain Fingerprints for No-reference Image Quality Assessment
• Unsupervised Many-to-Many Image-to-Image Translation Across Multiple Domains
• Attribute-guided Feature Extraction and Augmentation Robust Learning for Vehicle Re-identification
• Multi-Level Attention Pooling for Graph Neural Networks: Unifying Graph Representations with Multiple Localities
• Scale Calibrated Training: Improving Generalization of Deep Networks via Scale-Specific Normalization
• Understanding the Disharmony between Weight Normalization Family and Weight Decay: $\epsilon -$shifted $L_2$ Regularizer
• Multiple Style-Transfer in Real-Time
• Kunster -- AR Art Video Maker -- Real time video neural style transfer on mobile devices
• Deep Exposure Fusion with Deghosting via Homography Estimation and Attention Learning
• Unsupervised Anomaly Detection in MR Images using Multi-Contrast Information
• Unsupervised Facial Action Unit Intensity Estimation via Differentiable Optimization
• Fine-Grained Classroom Activity Detection from Audio with Neural Networks
• Critiquing-based Modeling of Subjective Preferences
• A Semi-Supervised Approach for Abnormal Event Prediction on Large Operational Network Time-Series Data
• Separate In Latent Space: Unsupervised Single Image Layer Separation
• Recognizing Instagram Filtered Images with Feature De-stylization
• Image Style Transfer and Content-Style Disentanglement
• Learning to adapt class-specific features across domains for semantic segmentation
• How Does BN Increase Collapsed Neural Network Filters?
• GANSeg: Learning to Segment by Unsupervised Hierarchical Image Generation
• Computationally Efficient Approaches for Image Style Transfer
• GANILLA: Generative Adversarial Networks for Image to Illustration Translation
• Batch Normalization with Enhanced Linear Transformation
• SubSpectral Normalization for Neural Audio Data Processing
• Deep Learning-Based MR Image Re-parameterization
• Multimodal Generation of Novel Action Appearances for Synthetic-to-Real Recognition of Activities of Daily Living
• Stereo Object Matching Network
• Universal Undersampled MRI Reconstruction
• Connections Between Pairs of Filters Improve the Accuracy of Convolutional Neural Networks
• A Domain Agnostic Normalization Layer for Unsupervised Adversarial Domain Adaptation
• PanoDR: Spherical Panorama Diminished Reality for Indoor Scenes
• Reducing the feature divergence of RGB and near-infrared images using Switchable Normalization
• Interactive Object Segmentation with Dynamic Click Transform
• Multi-views Embedding for Cattle Re-identification
• DepthwiseGANs: Fast Training Generative Adversarial Networks for Realistic Image Synthesis
• Excavate Condition-invariant Space by Intrinsic Encoder
• A Proof-of-Concept Study of Artificial Intelligence Assisted Contour Revision
• iButter: Neural Interactive Bullet Time Generator for Human Free-viewpoint Rendering
• Substitute Teacher Networks: Learning with Almost No Supervision
• Learning Discriminative Features Via Weights-biased Softmax Loss
• Luminance Attentive Networks for HDR Image and Panorama Reconstruction
• Spectral Image Visualization Using Generative Adversarial Networks
• Self-Supervised Sketch-to-Image Synthesis
• DeepStrip: High Resolution Boundary Refinement
• MI^2GAN: Generative Adversarial Network for Medical Image Domain Adaptation using Mutual Information Constraint
• Deep Controllable Backlight Dimming
• Video-to-Video Translation for Visual Speech Synthesis
• Learning Task-oriented Disentangled Representations for Unsupervised Domain Adaptation
• Regularized Adaptation for Stable and Efficient Continuous-Level Learning on Image Processing Networks
• A Novel Structured Natural Gradient Descent for Deep Learning
• Predictive Model for Assessment of Pathological Response of Colorectal Liver Metastases to Chemotherapy from CT Images
• Efficient Modelling Across Time of Human Actions and Interactions
• Legacy Photo Editing with Learned Noise Prior
• Semantic-driven Colorization
• Learning to Inpaint by Progressively Growing the Mask Regions
• ID-Unet: Iterative Soft and Hard Deformation for View Synthesis
• Unpaired Deep Learning for Accelerated MRI using Optimal Transport Driven CycleGAN
• Arbitrary Style Transfer using Graph Instance Normalization
• Two-Stream Appearance Transfer Network for Person Image Generation
• Learning Camera-Aware Noise Models
• Generation and Simulation of Yeast Microscopy Imagery with Deep Learning
• Examining Performance of Sketch-to-Image Translation Models with Multiclass Automatically Generated Paired Training Data
• STALP: Style Transfer with Auxiliary Limited Pairing
• ISCL: Interdependent Self-Cooperative Learning for Unpaired Image Denoising
• A Lightweight Music Texture Transfer System
• DINO: A Conditional Energy-Based GAN for Domain Translation
• GPU-Accelerated Mobile Multi-view Style Transfer
• Self-calibrated convolution towards glioma segmentation
• The Role of the Input in Natural Language Video Description
• CWY Parametrization: a Solution for Parallelized Optimization of Orthogonal and Stiefel Matrices
• Deep Automodulators
• Fusion of neural networks, for LIDAR-based evidential road mapping
• Condensed Composite Memory Continual Learning
• Efficient Semantic Image Synthesis via Class-Adaptive Normalization
• WeightAlign: Normalizing Activations by Weight Alignment
• 3D Conceptual Design Using Deep Learning
• Efficient Purely Convolutional Text Encoding
• Image Translation for Medical Image Generation -- Ischemic Stroke Lesions
• Image Translation via Fine-grained Knowledge Transfer
• Reliable Liver Fibrosis Assessment from Ultrasound using Global Hetero-Image Fusion and View-Specific Parameterization
• Unsupervised Deep Learning for MR Angiography with Flexible Temporal Resolution
• RS-Net: Regression-Segmentation 3D CNN for Synthesis of Full Resolution Missing Brain MRI in the Presence of Tumours
• Implicit Euler ODE Networks for Single-Image Dehazing
• Separable Batch Normalization for Robust Facial Landmark Localization with Cross-protocol Network Training
• AmoebaContact and GDFold: a new pipeline for rapid prediction of protein structures
• LEGAN: Disentangled Manipulation of Directional Lighting and Facial Expressions by Leveraging Human Perceptual Judgements
• LocalNorm: Robust Image Classification through Dynamically Regularized Normalization
• Unsupervised Learning of Depth and Depth-of-Field Effect from Natural Images with Aperture Rendering Generative Adversarial Networks
• Normalized Convolutional Neural Network
• Boosting Unconstrained Face Recognition with Auxiliary Unlabeled Data
• Exact Backpropagation in Binary Weighted Networks with Group Weight Transformations
• Deep Learning-based Frozen Section to FFPE Translation
• Deep Interactive Denoiser (DID) for X-Ray Computed Tomography
• Dynamic Matching Markets in Power Grid: Concepts and Solution using Deep Reinforcement Learning
• Learning joint lesion and tissue segmentation from task-specific hetero-modal datasets
• CT-Net: Complementary Transfering Network for Garment Transfer with Arbitrary Geometric Changes
• A Content Transformation Block For Image Style Transfer
• Peak Detection On Data Independent Acquisition Mass Spectrometry Data With Semisupervised Convolutional Transformers
• De-rendering the World's Revolutionary Artefacts
• Towards Low-Resource StarGAN Voice Conversion using Weight Adaptive Instance Normalization
• Segmentation overlapping wear particles with few labelled data and imbalance sample
• Blind Image Decomposition
• Instance-Level Meta Normalization
• A novel generative reverse net assisted evolution algorithm for expensive-computational optimizations
• What and Where to Translate: Local Mask-based Image-to-Image Translation
• New Interpretations of Normalization Methods in Deep Learning
• Simpler Certified Radius Maximization by Propagating Covariances
• Unpaired Single-Image Depth Synthesis with cycle-consistent Wasserstein GANs
• Global and Local Texture Randomization for Synthetic-to-Real Semantic Segmentation
• An Integrated Enhancement Solution for 24-hour Colorful Imaging
• Irregular Convolutional Auto-Encoder on Point Clouds
• Understanding the wiring evolution in differentiable neural architecture search
• Spatial Content Alignment For Pose Transfer
• Farkas layers: don't shift the data, fix the geometry
• An Internal Covariate Shift Bounding Algorithm for Deep Neural Networks by Unitizing Layers' Outputs
• Partial supervision for the FeTA challenge 2021
• Tackling Long-Tailed Relations and Uncommon Entities in Knowledge Graph Completion
• Head2HeadFS: Video-based Head Reenactment with Few-shot Learning
• Identifying Table Structure in Documents using Conditional Generative Adversarial Networks
• Interpretable Filter Learning Using Soft Self-attention For Raw Waveform Speech Recognition
• Non-Parametric Neural Style Transfer
• Bilinear Input Normalization for Neural Networks in Financial Forecasting
• NAPA: Neural Art Human Pose Amplifier
• NeuralMagicEye: Learning to See and Understand the Scene Behind an Autostereogram
• Neuronal Learning Analysis using Cycle-Consistent Adversarial Networks
• FDA: Feature Decomposition and Aggregation for Robust Airway Segmentation
• Picasso: A CUDA-based Library for Deep Learning over 3D Meshes
• Joint Deep Learning of Facial Expression Synthesis and Recognition