Jia-Bin HuangPublications [show all by date] / [show selected] / [show all by topic]Beyond Supervision LearningManifold Graph with Learned Prototypes for Semi-Supervised Image Classification A Closer Look at Few-shot Classification CrDoCo: Pixel-level Domain Transfer with Cross-Domain Consistency Unsupervised Representation Learning by Sorting Sequences Unsupervised Visual Representation Learning by Graph-based Consistent Constraints Progressive Representation Adaptation for Weakly Supervised Object Localization Towards Accurate and Robust Cross-Ratio based Gaze Trackers Through Learning From Simulation Correspondence and MatchingDF-Net: Unsupervised Joint Learning of Depth and Flow using Cross-Task Consistency Show, Match and Segment: Joint Learning of Semantic Matching and Object Co-segmentation Deep Semantic Matching with Foreground Detection and Cycle-Consistency DeepMVS: Learning Multi-View Stereopsis Semi-Supervised Learning for Optical Flow with Generative Adversarial Networks Video Object TrackingSAIL-VOS: Semantic Amodal Instance Level Video Object Segmentation - A Synthetic Dataset and Baselines Tracking Persons-of-Interest via Unsupervised Representation Adaptation VideoMatch: Matching based Video Object Segmentation Unsupervised Video Object Segmentation using Motion Saliency-Guided Spatio-Temporal Propagation MaskRNN: Instance Level Video Object Segmentation Robust Visual Tracking via Hierarchical Convolutional Features Adaptive Correlation Filters with Long-Term and Short-Term Memory
for Object Tracking Detecting Migrating Birds at Night Looking at PeopleLearning to Mitigate Scene Bias in Action Recognition iCAN: Instance-Centric Attention Network for Human-Object Interaction Detection Progressive Cyber-Human Intelligence for Social Good Semi-Automated Home-based Therapy for the Upper Extremity of Stroke Survivors HOMER: an interactive system for home based stroke rehabilitation Ensemble of Convolutional Neural Networks for Pose Estimation Computational Photography/VideographyGuided Image-to-Image Translation with Bi-Directional Feature Transformation DRIT++: Diverse Image-to-Image Translation via Disentangled Representations Learning Blind Video Temporal Consistency Fast and Accurate Image Super-Resolution with Deep Laplacian Pyramid Networks Joint Image Filtering with Deep Convolutional Networks Temporally Coherent Completion of Dynamic Video A Comparative Study for Single Image Blind Deblurring Single Image Super-Resolution from Transformed Self-Exemplars Image Completion using Planar Structure Guidance Transformation Guided Image Completion Image Recolorization For The Colorblind Information Preserving Color Transformation for Protanopia and Deuteranopia Adversarial Attacks and DefenseConnecting the Digital and Physical World: Improving the Robustness of Adversarial Attacks Detecting Adversarial Attacks on Neural Network Policies with Visual Foresight Fun stuffDeep Paper Gestalt Source Form: An Automated Crowdsourced Object Generator Multi-view Wire Art PhD ThesisVisual Analysis and Synthesis with Physically Grounded Constraints |
