Biomedical Imaging Center Deep Learning Journal Club
- Generative Adversarial Network
- Presenter: Qing [slides]
- Network In Network
- Presenter: Qing [slides]
- Learning Deep Features for Discriminative Localization
- Presenter: Hongming [slides]
- Human-level control through deep reinforcement learning
- Presenter: Qing [slides]
- NIPS 2016 Tutorial: Generative Adversarial Networks - Part I
- Presenter: Grant [slides]
- Understanding LSTM Networks
- Presenter: Hongming [slides]
- NIPS 2016 Tutorial: Generative Adversarial Networks - Part II
- Presenter: Grant
- Show and Tell: Lessons Learned from the 2015 MSCOCO Image Captioning Challenge
- NIPS 2016 Tutorial: Generative Adversarial Networks - Part III
- Presenter: Grant
- Brain2Image: Converting Brain signals into images
- Presenter: Yuanyuan [slides]
- MDNet: A Semantically and Visually Interpretable Medical Image Diagnosis Network
- Presenter: Qingsong [slides]
- Taskonomy: Disentangling Task Transfer Learning
- Presenter: Pingkun [slides]
- Lose the Views: Limited Angle CT Reconstruction via Implicit Sinogram Completion
- Presenter: Hongming [slides]
- Semantic Image Inpainting with Deep Generative Models
- Presenter: Lars [slides]
- Image reconstruction by domain-transform manifold learning
- Presenter: Jason
- Medical Image Synthesis with Deep Convolutional Adversarial Networks
- Presenter: Grant [slides]
- Deep Convolutional Framelet Denosing for Low-Dose CT via Wavelet Residual Network
- Presenter: Qingsong [slides]
- Deep learning with domain adaptation for accelerated projection‐reconstruction MR
- Presenter: Qing
- DeepSense: A Unified Deep Learning Framework for Time-Series Mobile Sensing Data Processing
- Presenter: YuanYuan [slides]
- Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
- Presenter: Lars [slides]
- CyCADA: Cycle-Consistent Adversarial Domain Adaptation
- Presenter: Pingkun [slides]
- Visualizing High-Dimensional Data Using t-SNE
- Presenter: Hongming [slides]
- Deep Lesion Graphs in the Wild: Relationship Learning and Organization of Significant Radiology Image Findings in a Diverse Large-scale Lesion Database
- Presenter: Grant
- Evaluating surgical skills from kinematic data using convolutional neural networks
- Presenter: Yuanyuan [slides]
- Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network
- Presenter: Hengtao [slides]
- Left-Right Comparative Recurrent Model for Stereo Matching
- Presenter: Xi [slides]
- Erase or Fill? Deep Joint Recurrent Rain Removal and Reconstruction in Videos
- Presenter: Lars [slides]
- High-Resolution Image Synthesis and Semantic Manipulation With Conditional GANs
- Presenter: Qing [slides]
- Large Scale GAN Training for High Fidelity Natural Image Synthesis
- Presenter: Hongming [slides]
- An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solution
- Presenter: Pingkun [slides]
- An Explainable Adversarial Robustness Metric for Deep Learning Neural Networks
- Presenter: Grant
- Rethinking the Faster R-CNN Architecture for Temporal Action Localization
- Presenter: Yuanyuan [slides]
- W2F: A Weakly-Supervised to Fully-Supervised Framework for Object Detection
- Presenter: Hengtao [slides]
- Compressed Sensing MRI Reconstruction Using a Generative Adversarial Network With a Cyclic Loss
- Presenter: Huidong [slides]
- Mask R-CNN
- Presenter: Xue [slides]
- Deep Extreme Cut: From Extreme Points to Object Segmentation
- Presenter: Xi [slides]
- Learning to Segment Every Thing
- Presenter: Hongming [slides]
- Deep Reinforcement Learning with Double Q-Learning
- Presenter: Qing [slides]
- SynSeg-Net: Synthetic Segmentation Without Target Modality Ground Truth
- Fully convolutional networks for semantic segmentation
- Presenter: Jason [slides]
- Iterative PET Image Reconstruction Using Convolutional Neural Network Representation
- Presenter: Lars [slides]
- Are GANs Created Equal? A Large-Scale Study
- Presenter: Grant
- Recognizing Brain States Using Deep Sparse Recurrent Neural Network
- Presenter: Yuanyuan [slides]
- Understanding deep learning requires rethinking generalization
- Presenter: Xue [slides]
- You only look once: Unified, real-time object detection
- Presenter: Hengtao [slides]
- Learning with a Wasserstein Loss
- Presenter: Xi [slides]
- Adapted Deep Embeddings: A Synthesis of Methods for k-Shot Inductive Transfer Learning
- Presenter: Fenglei
- Panoptic Feature Pyramid Networks
- Presenter: Xue [slides]
- GLoMo: Unsupervised Learning of Transferable Relational Graphs
- Presenter: Qing [slides]
- BourGAN: Generative Networks with Metric Embeddings
- Presenter: Hongming [slides]
- Wasserstein Introspective Neural Networks
- Presenter: Grant
- A Probabilistic U-Net for Segmentation of Ambiguous Images
- Presenter: Qikui
- Aggregated residual transformations for deep neural networks
- Presenter: Huidong [slides]
- Deep learning enables cross-modality super-resolution in fluorescence microscopy
- Presenter: Mengzhou [slides]
- Improved Semi-supervised Learning with GANs using Manifold Invariances
- Presenter: Fatir
- Self-Supervised Deep Active Accelerated MRI
- Presenter: Jason [slides]
- Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation
- Presenter: Xi [slides]
- Self-Attention Generative Adversarial Networks
- Presenter: Hengtao [slides]
- Fast, Accurate and Lightweight Super-Resolution with Neural Architecture Search
- Presenter: Fenglei
- Non-Local Neural Networks
- Presenter: Xue [slides]
- Deep image prior
- Presenter: Qing [slides]
- A Style-Based Generator Architecture for Generative Adversarial Networks
- Presenter: Hongming [slides]
- Group Normalization
- Presenter: Grant
- An integrated iterative annotation technique for easing neural network training in medical image analysis
- Presenter: Qikui
- Attention is all you need
- Presenter: Huidong [slides]
- Content-aware image restoration: pushing the limits of fluorescence microscopy
- Presenter: Mengzhou [slides]
- Deep Learning for Automated Contouring of Primary Tumor Volumes by MRI for Nasopharyngeal Carcinoma
- Presenter: Zhao
- Semi-Supervised Classification with Graph Convolutional Networks
- Presenter: Fatir [slides]
- Learning to Summarize Radiology Findings
- Presenter: Xi
- Auto-Keras: An Efficient Neural Architecture Search System
- Presenter: Hengtao
- Exploring Randomly Wired Neural Networks for Image Recognition
- Presenter: Fenglei
- DetNAS: Neural Architecture Search on Object Detection
- Presenter: Xue [slides]
- Single Path One-Shot Neural Architecture Search with Uniform Sampling
- Presenter: Qing [slides]
- Learning Discrete Structures for Graph Neural Networks
- Presenter: Hongming [slides]
- Learning to learn by gradient descent by gradient descent
- Presenter: Grant
- TensorMask: A Foundation for Dense Object Segmentation
- Presenter: Qikui
- Hierarchical Surface Prediction
- Presenter: Mengzhou [slides]
- MetaPruning: Meta Learning for Automatic Neural Network Channel Pruning
- Presenter: Huidong [slides]
- SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation
- Presenter: Zhao
- Dynamic Routing Between Capsules
- Presenter: Fatir
- Unsupervised deep learning for Bayesian brain MRI segmentation
- Presenter: Xi
- NAS-Bench-101: Towards Reproducible Neural Architecture Search
- Presenter: Hengtao
- Graph U-NET
- Presenter: Xue [slides]
- Self-Attention Graph Pooling
- Presenter: Qing [slides]
- HexaGAN: Generative Adversarial Nets for Real World Classification
- Presenter: Hongming [slides]
- Simplifying Graph Convolutional Networks
- Presenter: Qikui
- Disentangled Graph Convolutional Networks
- Presenter: Mengzhou [slides]
- Deep Learning-Based Image Segmentation on Multimodal Medical Imaging
- Presenter: Huidong [slides]
- Interpreting CNNs via Decision Trees
- Presenter: Jason [slides]
- Unsupervised Data Augmentation for Consistency Training
- Presenter: Fatir [slides]
- Weight Agnostic Neural Networks
- Presenter: Hengtao [slides]
- Elastic Boundary Projection for 3D Medical Imaging Segmentation
- Presenter: Xi [slides]
- Large Scale Adversarial Representation Learning
- Presenter: Hongming [slides]
- Deep Kalman Filtering Network for Video Compression Artifact Reduction
- Presenter: Qing [slides]
- Semi-Supervised Learning with Scarce Annotations
- Presenter: Qikui [slides]
- Compositional GAN: Learning Image-Conditional Binary Composition
- Presenter: Mengzhou [slides]
- DuDoNet: Dual Domain Network for CT Metal Artifact Reduction
- Presenter: Huidong [slides]
- Random Erasing Data Augmentation
- Presenter: Fatir [slides]
- Multiview 2D/3D Rigid Registration via a Point-Of-Interest Network for Tracking and Triangulation
- Presenter: Hengtao [slides]
- Self-supervised learning for medical image analysis using image context restoration
- Presenter: Xi [slides]
- Data-efficient image recognition with contrastive predictive coding
- Presenter: Hongming [slides]
- Video Enhancement with Task-Oriented Flow
- Presenter: Qing [slides]
- Edge-Labeling Graph Neural Network for Few-shot Learning
- Presenter: Qikui [slides]
- Revisiting Self-Supervised Visual Representation Learning
- Presenter: Xi [slides]
- Deep Information Theoretic Registration
- Presenter: Hengtao [slides]
- Matching Networks for One Shot Learning
- Presenter: Pingkun [slides]
- Generating Classification Weights with GNN Denoising Autoencoders for
Few-Shot Learning
- Presenter: Hanqing [slides]
- Shuffle and Learn: Unsupervised Learning using Temporal Order Verification
- Presenter: Huidong
- Colorful Image Colorization
- Presenter: Fatir
- Self-Supervised Spatio-Temporal Representation Learning for Videos by Predicting Motion and Appearance Statistics
- Presenter: Qing
- Neural Ordinary Differential Equations
- Presenter: Fenglei
- Learning Features by Watching Objects Move
- Presenter: Chris
- Discriminative Unsupervised Feature Learning with Convolutional Neural Networks
- Presenter: Huihua
- Networks for Joint Affine and Non-parametric Image Registration
- Presenter: Hengtao [slides]
- SCOPS: Self-Supervised Co-Part Segmentation
- Presenter: Xi
- beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework
- Presenter: Xi
- Towards a Definition of Disentangled Representations
- Presenter: Hengtao [slides]
- Tutorial on Variational Autoencoders
- Presenter: Hongming [slides]
- LOGAN: Latent Optimisation for Generative Adversarial Networks
- Presenter: Qing [slides]
- Multi-source Domain Adaptation for Semantic Segmentation
- Presenter: Huihua [slides]
- Momentum Contrast for Unsupervised Visual Representation Learning
- Presenter: Hanqing [slides]
- Deep Self-Learning From Noisy Labels
- Presenter: Fatir [slides]
- S4L: Self-Supervised Semi-Supervised Learning
- Presenter: Chris [slides]
- Attention Augmented Convolutional Networks
- Presenter: Qiyun [slides]
- Many Task Learning With Task Routing
- Presenter: Huidong [slides]
- Distillation-Based Training for Multi-Exit Architectures
- Presenter: Mengzhou
- ADN: Artifact Disentanglement Network for Unsupervised Metal Artifact Reduction
- Presenter: Chuang [slides]
- Putting An End to End-to-End: Gradient-Isolated Learning of Representations
- Presenter: Hongming [slides]
- Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations
- Presenter: Hanqing [slides]
- SinGAN: Learning a Generative Model From a Single Natural Image
- Presenter: Qing [slides]
- ShapeMask: Learning to Segment Novel Objects by Refining Shape Prior
- Presenter: Xi [slides]
- Models genesis: Generic autodidactic models for 3D medical image analysis
- Presenter: Hengtao [slides]
- What’s Hidden in a Randomly Weighted Neural Network?
- Presenter: Huihua [slides]
- Unsupervised Domain Adaptation through Self-Supervision
- Presenter: Fatir
- A simple framework for contrastive learning of visual representations
- Presenter: Chris [slides]
- UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation
- Presenter: Chuang [slides]
- SROBB: Targeted Perceptual Loss for Single Image Super-Resolution
- Presenter: Mengzhou [slides]
- Stacked spatio-temporal graph convolutional networks for action segmentation
- Presenter: Huidong [slides]
- Stacked Capsule Autoencoders
- Presenter: Hongming [slides]
- Disentangling Propagation and Generation for Video Prediction
- Presenter: Yuanyuan [slides]
- Verification of non-linear specifications for neural networks
- Presenter: Hanqing [slides]
- Learning Correspondence from the Cycle-Consistency of Time
- Presenter: Qing [slides]
- Large-Scale Screening of COVID-19 from Community Acquired Pneumonia using Infection Size-Aware Classification
- Presenter: Xiaodong [slides]
- SkipNet: Learning Dynamic Routing in Convolutional Networks
- Presenter: Qiyun [slides]
- Adversarial Continual Learning
- Presenter: Xi [slides]
- ResNeSt: Split-Attention Networks
- Presenter: Hengtao [slides]
- Drop an octave: Reducing spatial redundancy in convolutional neural networks with octave convolution
- Presenter: Huidong [slides]
- Federated Adversarial Domain Adaptation
- Presenter: Hongming [slides]
- Decoupling Representation and Classifier for Long-Tailed Recognition
- Presenter: Fatir
- Improving Semantic Segmentation via Self-Training
- Presenter: Chris [slides]
- Gated-SCNN: Gated Shape CNNs for Semantic Segmentation
- Presenter: Jiajin [slides]
- Learning To Classify Images Without Labels
- Presenter: Chuang [slides]
- Evaluating Robustness of Deep Image Super-Resolution Against Adversarial Attacks
- Presenter: Mengzhou [slides]
- Adversarial Training and Robustness for Multiple Perturbations
- Presenter: Hanqing [slides]
- How Does Batch Normalization Help Optimization?
- Presenter: Jason [slides]
- Non-Local ConvLSTM for Video Compression Artifact Reduction
- Presenter: Qing [slides]
- Domain specific cues improve robustness of deep learning based segmentation of CT volumes
- Presenter: Xiaodong [slides]
- Time-series Generative Adversarial Networks
- Presenter: Yuanyuan [slides]
- On instabilities of deep learning in image reconstruction and the potential costs of AI
- Presenter: Weiwen
- YOLACT Real-time Instance Segmentation
- Presenter: Xi [slides]
- CNN-generated images are surprisingly easy to spot... for now
- Presenter: Hengtao [slides]
- Big Self-Supervised Models are Strong Semi-Supervised Learners
- Presenter: Hongming [slides]
- Contrastive Learning for Unpaired Image-to-Image Translation
- Presenter: Huidong [slides]
- The Limitations of Deep Learning in Adversarial Settings
- Presenter: Hengtao [slides]
- Distillation as a Defense to Adversarial Perturbations against Deep Neural Networks
- Presenter: Mengzhou[slides]
-
On the Robustness of Semantic Segmentation Models to Adversarial Attacks
- Presenter: Xi [slides]
-
- Presenter: Nneka [slides]
-
Adversarial Discriminative Domain Adaptation
- Presenter: Nathan [slides]
-
Simultaneous deep transfer across domains and tasks
- Presenter: Diego [slides]
-
Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization
- Presenter: Xinrui
-
Understanding Adversarial Examples from the Mutual Influence of Images and perturbations
- Presenter: Fatir
-
Learning Representations for Time Series Clustering
- Presenter: Chris
-
Robust Learning Through Cross-Task Consistency
- Presenter: Jiajin
-
Modeling Biological Immunity to Adversarial Examples
- Presenter: Jason
-
Soft Labels for Ordinal Regression
- Presenter: Xiaodong
-
VL-BERT: Pre-training of Generic Visual-Linguistic Representations
- Presenter: Chuang
-
The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks
- Presenter: Fenglei
-
Towards Deep Learning Models Resistant to Adversarial Attacks
- Presenter: Dr. Cong
-
Towards Evaluating the Robustness of Neural Networks
- Presenter: Hanqing
-
Adversarial Examples Improve Image Recognition
- Presenter: Weiwen