ECCV 2018 accepted papers on Label-efficient and robust learning
103 of the 776 papers accepted at ECCV 2018 (13.3%) are on Label-efficient and robust learning, compared with 10.1% at ECCV 2016. They fall into 8 topics.
See ECCV 2018 in the interactive map
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 · 59 papers
- Deep Clustering for Unsupervised Learning of Visual Features (2,745 citations)
- Beyond Part Models: Person Retrieval with Refined Part Pooling (and A Strong Convolutional Baseline) (2,694 citations)
- Learning to Navigate for Fine-grained Classification (649 citations)
- Part-Aligned Bilinear Representations for Person Re-Identification (577 citations)
- Open Set Learning with Counterfactual Images (512 citations)
- Deep Metric Learning with Hierarchical Triplet Loss (475 citations)
- Mancs: A Multi-task Attentional Network with Curriculum Sampling for Person Re-identification (421 citations)
- Textual Explanations for Self-Driving Vehicles (392 citations)
- Hierarchical Bilinear Pooling for Fine-Grained Visual Recognition (368 citations)
- Facial Expression Recognition with Inconsistently Annotated Datasets (362 citations)
- Person Re-identification with Deep Similarity-Guided Graph Neural Network (341 citations)
- Interpretable Basis Decomposition for Visual Explanation (276 citations)
and 47 more in the interactive map.
Domain adaptation and generalization · 24 papers
- Unsupervised Domain Adaptation for Semantic Segmentation via Class-Balanced Self-Training (1,483 citations)
- Two at Once: Enhancing Learning and Generalization Capacities via IBN-Net (967 citations)
- Deep Domain Generalization via Conditional Invariant Adversarial Networks (634 citations)
- Open Set Domain Adaptation by Backpropagation (607 citations)
- Generalizing A Person Retrieval Model Hetero- and Homogeneously (508 citations)
- Partial Adversarial Domain Adaptation (493 citations)
- DeepJDOT: Deep Joint Distribution Optimal Transport for Unsupervised Domain Adaptation (469 citations)
- Transferring GANs: generating images from limited data (287 citations)
- Unsupervised Person Re-identification by Deep Learning Tracklet Association (256 citations)
- DCAN: Dual Channel-wise Alignment Networks for Unsupervised Scene Adaptation (252 citations)
- Domain Adaptation through Synthesis for Unsupervised Person Re-identification (231 citations)
- Effective Use of Synthetic Data for Urban Scene Semantic Segmentation (164 citations)
and 12 more in the interactive map.
Semi-supervised learning · 7 papers
- Deep Co-Training for Semi-Supervised Image Recognition (554 citations)
- Transductive Semi-Supervised Deep Learning using Min-Max Features (224 citations)
- Semi-Supervised Deep Learning with Memory (65 citations)
- HybridNet: Classification and Reconstruction Cooperation for Semi-Supervised Learning (51 citations)
- Semi-Supervised Generative Adversarial Hashing for Image Retrieval (40 citations)
- Transductive Centroid Projection for Semi-supervised Large-scale Recognition (34 citations)
- SaaS: Speed as a Supervisor for Semi-supervised Learning (19 citations)
Zero-/few-shot and meta-learning · 6 papers
- Multi-modal Cycle-consistent Generalized Zero-Shot Learning (396 citations)
- A Zero-Shot Framework for Sketch based Image Retrieval (204 citations)
- Learning Class Prototypes via Structure Alignment for Zero-Shot Recognition (143 citations)
- Dynamic Conditional Networks for Few-Shot Learning (75 citations)
- Choose Your Neuron: Incorporating Domain Knowledge through Neuron-Importance (44 citations)
- Selective Zero-Shot Classification with Augmented Attributes (27 citations)
Knowledge distillation · 3 papers
Self-supervised and contrastive · 2 papers
- Fighting Fake News: Image Splice Detection via Learned Self-Consistency (481 citations)
- Conditional Prior Networks for Optical Flow (34 citations)
