CVPR 2018 accepted papers on Label-efficient and robust learning
150 of the 979 papers accepted at CVPR 2018 (15.3%) are on Label-efficient and robust learning, compared with 11.1% at CVPR 2017. They fall into 9 topics.
See CVPR 2018 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 · 83 papers
- Unsupervised Feature Learning via Non-Parametric Instance Discrimination (3,635 citations)
- Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics (3,007 citations)
- CosFace: Large Margin Cosine Loss for Deep Face Recognition (2,951 citations)
- Person Transfer GAN to Bridge Domain Gap for Person Re-Identification (2,058 citations)
- Harmonious Attention Network for Person Re-Identification (1,430 citations)
- Taskonomy: Disentangling Task Transfer Learning (1,182 citations)
- Learning Deep Models for Face Anti-Spoofing: Binary or Auxiliary Supervision (725 citations)
- Mask-Guided Contrastive Attention Model for Person Re-Identification (705 citations)
- Camera Style Adaptation for Person Re-Identification (658 citations)
- The Power of Ensembles for Active Learning in Image Classification (645 citations)
- Multi-Level Factorisation Net for Person Re-Identification (564 citations)
- Features for Multi-Target Multi-Camera Tracking and Re-Identification (551 citations)
and 71 more in the interactive map.
Domain adaptation and generalization · 31 papers
- Maximum Classifier Discrepancy for Unsupervised Domain Adaptation (2,280 citations)
- Learning to Adapt Structured Output Space for Semantic Segmentation (1,793 citations)
- Domain Generalization With Adversarial Feature Learning (1,271 citations)
- Image-Image Domain Adaptation With Preserved Self-Similarity and Domain-Dissimilarity for Person Re-Identification (1,045 citations)
- Transferable Joint Attribute-Identity Deep Learning for Unsupervised Person Re-Identification (702 citations)
- Generate to Adapt: Aligning Domains Using Generative Adversarial Networks (680 citations)
- Image to Image Translation for Domain Adaptation (571 citations)
- Collaborative and Adversarial Network for Unsupervised Domain Adaptation (554 citations)
- Large Scale Fine-Grained Categorization and Domain-Specific Transfer Learning (525 citations)
- Learning From Synthetic Data: Addressing Domain Shift for Semantic Segmentation (520 citations)
- Importance Weighted Adversarial Nets for Partial Domain Adaptation (482 citations)
- Partial Transfer Learning With Selective Adversarial Networks (462 citations)
and 19 more in the interactive map.
Zero-/few-shot and meta-learning · 18 papers
- Dynamic Few-Shot Visual Learning Without Forgetting (1,151 citations)
- Feature Generating Networks for Zero-Shot Learning (1,080 citations)
- Low-Shot Learning From Imaginary Data (772 citations)
- Few-Shot Image Recognition by Predicting Parameters From Activations (583 citations)
- Generalized Zero-Shot Learning via Synthesized Examples (519 citations)
- A Generative Adversarial Approach for Zero-Shot Learning From Noisy Texts (438 citations)
- Learning to Compare: Relation Network for Few-Shot Learning (372 citations)
- Multi-Label Zero-Shot Learning With Structured Knowledge Graphs (335 citations)
- Transductive Unbiased Embedding for Zero-Shot Learning (233 citations)
- Preserving Semantic Relations for Zero-Shot Learning (184 citations)
- Discriminative Learning of Latent Features for Zero-Shot Recognition (177 citations)
- Low-Shot Learning With Large-Scale Diffusion (133 citations)
and 6 more in the interactive map.
Self-supervised and contrastive · 6 papers
- The Unreasonable Effectiveness of Deep Features as a Perceptual Metric (13,855 citations)
- Self-Supervised Adversarial Hashing Networks for Cross-Modal Retrieval (495 citations)
- Boosting Self-Supervised Learning via Knowledge Transfer (323 citations)
- Cross-Domain Self-Supervised Multi-Task Feature Learning Using Synthetic Imagery (200 citations)
- Improvements to Context Based Self-Supervised Learning (139 citations)
- Self-Supervised Feature Learning by Learning to Spot Artifacts (120 citations)
Noisy labels · 5 papers
- Joint Optimization Framework for Learning With Noisy Labels (728 citations)
- CleanNet: Transfer Learning for Scalable Image Classifier Training With Label Noise (497 citations)
- Iterative Learning With Open-Set Noisy Labels (320 citations)
- Deep Unsupervised Saliency Detection: A Multiple Noisy Labeling Perspective (195 citations)
- Learning From Noisy Web Data With Category-Level Supervision (26 citations)
Semi-supervised learning · 3 papers
- Learning to Localize Sound Source in Visual Scenes (349 citations)
- Smooth Neighbors on Teacher Graphs for Semi-Supervised Learning (279 citations)
- Improving Landmark Localization With Semi-Supervised Learning (184 citations)
OOD and anomaly detection · 2 papers
Knowledge distillation · 1 paper
- Deep Mutual Learning (1,819 citations)
Test-time adaptation · 1 paper
- DiverseNet: When One Right Answer Is Not Enough (21 citations)
