Shengfeng He

Publications

• Blind Image Denoising via Dynamic Dual Learning
• Video Snapshot: Single Image Motion Expansion via Invertible Motion Embedding
• Multi-View Face Synthesis via Progressive Face Flow
• Erratum to “Multi-View Face Synthesis via Progressive Face Flow”
• Pro-PULSE: Learning Progressive Encoders of Latent Semantics in GANs for Photo Upsampling
• Efficient Exploration in Crowds by Coupling Navigation Controller and Exploration Planner
• Invertible Grayscale via Dual Features Ensemble
• Deep Multiview Clustering via Iteratively Self-Supervised Universal and Specific Space Learning
• CrowdGAN: Identity-Free Interactive Crowd Video Generation and Beyond
• Holistically Associated Transductive Zero-Shot Learning
• DSDNet: Toward single image deraining with self-paced curricular dual stimulations
• Parsing-Conditioned Anime Translation: A New Dataset and Method
• Editing Out-of-Domain GAN Inversion via Differential Activations
• Weakly supervised segmentation via instance-aware propagation
• Pose- and Attribute-consistent Person Image Synthesis
• Monocular Depth Estimation for Glass Walls with Context: A New Dataset and Method
• Self-Supervised Matting-Specific Portrait Enhancement and Generation
• Disentangling Multi-view Representations Beyond Inductive Bias
• DAOT: Domain-Agnostically Aligned Optimal Transport for Domain-Adaptive Crowd Counting
• NPF-200: A Multi-Modal Eye Fixation Dataset and Method for Non-Photorealistic Videos
• Appearance-preserved Portrait-to-anime Translation via Proxy-guided Domain Adaptation
• Fully Deformable Network for Multiview Face Image Synthesis
• Reducing Spatial Labeling Redundancy for Active Semi-Supervised Crowd Counting
• Single-View View Synthesis with Self-rectified Pseudo-Stereo
• Layout generation as intermediate action sequence prediction
• SCANet: Self-paced semi-curricular attention network for non-homogeneous image dehazing
• Contextual-Assisted Scratched Photo Restoration
• TranSiam: Aggregating multi-modal visual features with locality for medical image segmentation
• Smart Scribbles for Image Matting
• Make Your Own Sprites: Aliasing-Aware and Cell-Controllable Pixelization
• SeqSeg: A sequential method to achieve nasopharyngeal carcinoma segmentation free from background dominance
• Self-Supervised Video Representation Learning by Uncovering Spatio-Temporal Statistics
• Fast scene labeling via structural inference
• Learning invariant and uniformly distributed feature space for multi-view generation?
• FormNet: Formatted Learning for Image Restoration
• Mask-Guided Deformation Adaptive Network for Human Parsing
• Image captioning via semantic element embedding
• Two-stage Photograph Cartoonization via Line Tracing
• Invertible Grayscale with Sparsity Enforcing Priors
• Deep Pixel-Level Matching via Attention for Video Co-Segmentation
• Edge Distraction-aware Salient Object Detection
• Example‐Based Colourization Via Dense Encoding Pyramids
• Coupled Rain Streak and Background Estimation via Separable Element-Wise Attention
• Crowd Counting Via Cross-Stage Refinement Networks
• Fast User-Guided Single Image Reflection Removal via Edge-Aware Cascaded Networks
• Transductive Zero-Shot Action Recognition via Visually Connected Graph Convolutional Networks
• Boundary-Aware RGBD Salient Object Detection With Cross-Modal Feature Sampling
• Unsupervised Domain Adaptation via Importance Sampling
• Mask-ShadowNet: Toward Shadow Removal via Masked Adaptive Instance Normalization
• Few-Shot Breast Cancer Metastases Classification via Unsupervised Cell Ranking
• Real-time salient object detection with a minimum spanning tree
• Exemplar-driven top-down saliency detection via deep association
• Oriented object proposals
• Fast Weighted Histograms for Bilateral Filtering and Nearest Neighbor Searching
• Interactive Hierarchical Object Proposals
• Efficient image super-resolution integration
• DeshadowNet: A Multi-context Embedding Deep Network for Shadow Removal
• Consistent stereo image editing
• TENet: Triple Excitation Network for Video Salient Object Detection
• Saliency detection with flash and no-flash image pairs
• Proposal-Driven Segmentation for Videos
• Saliency-guided color-to-gray conversion using region-based optimization
• Synthetic controllable turbulence using robust second vorticity confinement
• SuperCNN: A Superpixelwise Convolutional Neural Network for Salient Object Detection
• An efficient adaptive vortex particle method for real-time smoke simulation
• Real-time smoke simulation with improved turbulence by spatial adaptive vorticity confinement
• Visual tracking via locality sensitive histograms
• A Tool-Free Calibration Method for Turntable-Based 3D Scanning Systems
• Deep binocular tone mapping
• Joint Face Hallucination and Deblurring via Structure Generation and Detail Enhancement
• Stereo Object Proposals
• RGBD Salient Object Detection via Deep Fusion
• Egocentric Temporal Action Proposals
• $L_{0}$ -Regularized Image Downscaling
• Delving into Salient Object Subitizing and Detection
• Learning to Hallucinate Face Images via Component Generation and Enhancement
• FormResNet: Formatted Residual Learning for Image Restoration
• Keyword-driven image captioning via Context-dependent Bilateral LSTM
• Stylizing face images via multiple exemplars
• Joint Image Denoising and Disparity Estimation via Stereo Structure PCA and Noise-Tolerant Cost
• Egocentric Hand Detection Via Dynamic Region Growing
• Robust Object Tracking via Locality Sensitive Histograms
• Deformable Object Tracking With Gated Fusion
• Age estimation via attribute-region association
• Exploring Duality in Visual Question-Driven Top-Down Saliency
• Learning Long-Term Structural Dependencies for Video Salient Object Detection
• Real-Time Hierarchical Supervoxel Segmentation via a Minimum Spanning Tree
• Monocular Depth Estimation for Glass Walls With Context: A New Dataset and Method
• Reference-based screentone transfer via pattern correspondence and regularization
• Delving deep into pixelized face recovery and defense
• SINet: A scale-insensitive convolutional neural network for fast vehicle detection
• Learning transferable perturbations for image captioning
• Deep unsupervised pixelization
• Projecting your view attentively: Monocular road scene layout estimation via cross-view transformation
• A simple data mixing prior for improving self-supervised learning
• Reciprocal transformations for unsupervised video object segmentation
• Faithful extreme rescaling via generative prior reciprocated invertible representations
• Single image reflection removal beyond linearity
• Coherence and identity learning for arbitrary-length face video generation
• Learning from the master: Distilling cross-modal advanced knowledge for lip reading
• Towards a smaller student: Capacity dynamic distillation for efficient image retrieval
• Self-supervised spatio-temporal representation learning for videos by predicting motion and appearance statistics
• Fine-grained domain adaptive crowd counting via point-derived segmentation
• Surgical activity triplet recognition via triplet disentanglement
• Curricular contrastive regularization for physics-aware single image dehazing
• Where is my spot? Few-shot image generation via latent subspace optimization
• CIRI: Curricular inactivation for residue-aware one-shot video inpainting
• Deep video demoireing via compact invertible dyadic decomposition
• Diffuse3D: Wide-angle 3D photography via bilateral diffusion
• RIGID: Recurrent GAN inversion and editing of real face videos
• Background matting via recursive excitation
• DR-FER: Discriminative and Robust Representation Learning for Facial Expression Recognition
• High-resolution face swapping via latent semantics disentanglement
• Shunted self-attention via multi-scale token aggregation
• Delving deep into many-to-many attention for few-shot video object segmentation
• Differentiated learning for multi-modal domain adaptation
• Identity-Aware Variational Autoencoder for Face Swapping
• Joint shape matching for overlapping cytoplasm segmentation in cervical smear images
• Visualizing the invisible: Occluded vehicle segmentation and recovery
• Discovering interpretable latent space directions of gans beyond binary attributes
• Context-aware and scale-insensitive temporal repetition counting
• Spatially-invariant style-codes controlled makeup transfer
• From continuity to editability: Inverting GANs with consecutive images
• From contexts to locality: Ultra-high resolution image segmentation via locality-aware contextual correlation
• Co-advise: Cross inductive bias distillation
• Glance to count: Learning to rank with anchors for weakly-supervised crowd counting
• Active matting
• GDFace: Gated deformation for multi-view face image synthesis
• Don't hit me! glass detection in real-world scenes
• Delving into Multi-illumination Monocular Depth Estimation: A New Dataset and Method
• Learning an Interpretable Stylized Subspace for 3D-aware Animatable Artforms
• Make your own sprites: Aliasing-aware and cell-controllable pixelization
• Mask-guided deformation adaptive network for human parsing
• Smart scribbles for image matting
• Invertible grayscale with sparsity enforcing priors
• Monocular BEV Perception of Road Scenes Via Front-to-Top View Projection
• DiTMoS: Delving into diverse tiny-model selection on microcontrollers
• Unifying Global-Local Representations in Salient Object Detection With Transformers
• Hierarchical damage correlations for old photo restoration
• Delving into multimodal prompting for fine-grained visual classification
• Granular3D: Delving into multi-granularity 3D scene graph prediction
• DreamAnime: Learning Style-Identity Textual Disentanglement for Anime and Beyond
• Delving into Important Samples of Semi-Supervised Old Photo Restoration: A New Dataset and Method
• Learning Nighttime Semantic Segmentation the Hard Way
• 3D Snapshot: Invertible Embedding of 3D Neural Representations in a Single Image
• Triadic temporal-semantic alignment for weakly-supervised video moment retrieval
• Data from: Delving into Multi-illumination Monocular Depth Estimation: A New Dataset and Method
• Data from: Delving into Multi-illumination Monocular Depth Estimation: A New Dataset and Method
• G2Face: High-Fidelity Reversible Face Anonymization via Generative and Geometric Priors
• Modality-Aware Discriminative Fusion Network for Integrated Analysis of Brain Imaging Genomics