CVPR 2019 accepted papers on Label-efficient and robust learning
206 of the 1,294 papers accepted at CVPR 2019 (15.9%) are on Label-efficient and robust learning, compared with 15.3% at CVPR 2018. They fall into 8 topics.
See CVPR 2019 in the interactive map
Submitting to CVPR? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Recognition and long-tail · 90 papers
- ArcFace: Additive Angular Margin Loss for Deep Face Recognition (7,708 citations)
- Class-Balanced Loss Based on Effective Number of Samples (2,852 citations)
- Large-Scale Long-Tailed Recognition in an Open World (1,159 citations)
- Joint Discriminative and Generative Learning for Person Re-Identification (869 citations)
- Multi-Similarity Loss With General Pair Weighting for Deep Metric Learning (847 citations)
- Unsupervised Embedding Learning via Invariant and Spreading Instance Feature (601 citations)
- OCGAN: One-Class Novelty Detection Using GANs With Constrained Latent Representations (571 citations)
- Knockoff Nets: Stealing Functionality of Black-Box Models (537 citations)
- Learning to Reduce Dual-Level Discrepancy for Infrared-Visible Person Re-Identification (512 citations)
- Classification-Reconstruction Learning for Open-Set Recognition (473 citations)
- SpotTune: Transfer Learning Through Adaptive Fine-Tuning (472 citations)
- Why ReLU Networks Yield High-Confidence Predictions Far Away From the Training Data and How to Mitigate the Problem (443 citations)
and 78 more in the interactive map.
Domain adaptation and generalization · 47 papers
- Contrastive Adaptation Network for Unsupervised Domain Adaptation (967 citations)
- Domain Generalization by Solving Jigsaw Puzzles (862 citations)
- Taking a Closer Look at Domain Shift: Category-Level Adversaries for Semantics Consistent Domain Adaptation (846 citations)
- Bidirectional Learning for Domain Adaptation of Semantic Segmentation (715 citations)
- Invariance Matters: Exemplar Memory for Domain Adaptive Person Re-Identification (709 citations)
- Sliced Wasserstein Discrepancy for Unsupervised Domain Adaptation (618 citations)
- Learning From Synthetic Data for Crowd Counting in the Wild (616 citations)
- Universal Domain Adaptation (501 citations)
- Progressive Feature Alignment for Unsupervised Domain Adaptation (482 citations)
- Characterizing and Avoiding Negative Transfer (478 citations)
- Domain-Specific Batch Normalization for Unsupervised Domain Adaptation (437 citations)
- Sim-To-Real via Sim-To-Sim: Data-Efficient Robotic Grasping via Randomized-To-Canonical Adaptation Networks (425 citations)
and 35 more in the interactive map.
Zero-/few-shot and meta-learning · 33 papers
- Meta-Learning With Differentiable Convex Optimization (1,317 citations)
- Meta-Transfer Learning for Few-Shot Learning (1,294 citations)
- Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders (760 citations)
- Revisiting Local Descriptor Based Image-To-Class Measure for Few-Shot Learning (674 citations)
- F-VAEGAN-D2: A Feature Generating Framework for Any-Shot Learning (609 citations)
- Task Agnostic Meta-Learning for Few-Shot Learning (490 citations)
- Finding Task-Relevant Features for Few-Shot Learning by Category Traversal (393 citations)
- Leveraging the Invariant Side of Generative Zero-Shot Learning (343 citations)
- Attentive Region Embedding Network for Zero-Shot Learning (325 citations)
- Generating Classification Weights With GNN Denoising Autoencoders for Few-Shot Learning (307 citations)
- Generative Dual Adversarial Network for Generalized Zero-Shot Learning (262 citations)
- Image Deformation Meta-Networks for One-Shot Learning (255 citations)
and 21 more in the interactive map.
Semi-supervised learning · 9 papers
- Label Propagation for Deep Semi-Supervised Learning (690 citations)
- Semi-Supervised Transfer Learning for Image Rain Removal (476 citations)
- Data Augmentation Using Learned Transformations for One-Shot Medical Image Segmentation (459 citations)
- Unsupervised Deep Tracking (426 citations)
- Mutual Learning of Complementary Networks via Residual Correction for Improving Semi-Supervised Classification (51 citations)
- KE-GAN: Knowledge Embedded Generative Adversarial Networks for Semi-Supervised Scene Parsing (50 citations)
- Enhancing TripleGAN for Semi-Supervised Conditional Instance Synthesis and Classification (33 citations)
- Tangent-Normal Adversarial Regularization for Semi-Supervised Learning (29 citations)
- Joint Representative Selection and Feature Learning: A Semi-Supervised Approach (4 citations)
Knowledge distillation · 9 papers
- Relational Knowledge Distillation (1,698 citations)
- Variational Information Distillation for Knowledge Transfer (669 citations)
- Distilling Object Detectors With Fine-Grained Feature Imitation (466 citations)
- Knowledge Distillation via Instance Relationship Graph (323 citations)
- Progressive Teacher-Student Learning for Early Action Prediction (141 citations)
- Learning Metrics From Teachers: Compact Networks for Image Embedding (129 citations)
- Distilled Person Re-Identification: Towards a More Scalable System (111 citations)
- Student Becoming the Master: Knowledge Amalgamation for Joint Scene Parsing, Depth Estimation, and More (78 citations)
- Unifying Heterogeneous Classifiers With Distillation (60 citations)
Noisy labels · 9 papers
- Graph Convolutional Label Noise Cleaner: Train a Plug-And-Play Action Classifier for Anomaly Detection (594 citations)
- Probabilistic End-To-End Noise Correction for Learning With Noisy Labels (454 citations)
- Learning to Learn From Noisy Labeled Data (354 citations)
- Learning From Noisy Labels by Regularized Estimation of Annotator Confusion (248 citations)
- Unequal-Training for Deep Face Recognition With Long-Tailed Noisy Data (134 citations)
- MetaCleaner: Learning to Hallucinate Clean Representations for Noisy-Labeled Visual Recognition (114 citations)
- Noise-Tolerant Paradigm for Training Face Recognition CNNs (85 citations)
- Label-Noise Robust Generative Adversarial Networks (68 citations)
- Weakly Supervised Image Classification Through Noise Regularization (35 citations)
Self-supervised and contrastive · 8 papers
- Revisiting Self-Supervised Visual Representation Learning (690 citations)
- SelFlow: Self-Supervised Learning of Optical Flow (339 citations)
- Self-Supervised GANs via Auxiliary Rotation Loss (330 citations)
- Learning to Learn How to Learn: Self-Adaptive Visual Navigation Using Meta-Learning (256 citations)
- Self-Supervised Representation Learning by Rotation Feature Decoupling (229 citations)
- Self-Supervised Convolutional Subspace Clustering Network (173 citations)
- Self-Supervised Representation Learning From Videos for Facial Action Unit Detection (116 citations)
- Discovering Visual Patterns in Art Collections With Spatially-Consistent Feature Learning (100 citations)
Test-time adaptation · 1 paper
- Real-Time Self-Adaptive Deep Stereo (294 citations)
