CVPR 2016 accepted papers on Label-efficient and robust learning
82 of the 643 papers accepted at CVPR 2016 (12.8%) are on Label-efficient and robust learning. They fall into 6 topics.
See CVPR 2016 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 · 65 papers
- DeepFashion: Powering Robust Clothes Recognition and Retrieval With Rich Annotations (1,931 citations)
- Deep Metric Learning via Lifted Structured Feature Embedding (1,740 citations)
- Towards Open Set Deep Networks (1,699 citations)
- Cross-Stitch Networks for Multi-Task Learning (1,457 citations)
- Person Re-Identification by Multi-Channel Parts-Based CNN With Improved Triplet Loss Function (1,300 citations)
- CNN-RNN: A Unified Framework for Multi-Label Image Classification (1,236 citations)
- Learning Deep Representation for Imbalanced Classification (1,044 citations)
- Learning Deep Feature Representations With Domain Guided Dropout for Person Re-Identification (990 citations)
- The MegaFace Benchmark: 1 Million Faces for Recognition at Scale (978 citations)
- Patch-Based Convolutional Neural Network for Whole Slide Tissue Image Classification (904 citations)
- Compact Bilinear Pooling (896 citations)
- Deep Relative Distance Learning: Tell the Difference Between Similar Vehicles (821 citations)
and 53 more in the interactive map.
Domain adaptation and generalization · 6 papers
- Learning Multi-Domain Convolutional Neural Networks for Visual Tracking (2,839 citations)
- Learning Cross-Domain Landmarks for Heterogeneous Domain Adaptation (200 citations)
- Learning Attributes Equals Multi-Source Domain Generalization (175 citations)
- Constrained Deep Transfer Feature Learning and Its Applications (37 citations)
- The Multiverse Loss for Robust Transfer Learning (25 citations)
- Multivariate Regression on the Grassmannian for Predicting Novel Domains (17 citations)
Zero-/few-shot and meta-learning · 6 papers
- Latent Embeddings for Zero-Shot Classification (702 citations)
- Zero-Shot Learning via Joint Latent Similarity Embedding (362 citations)
- Semi-Supervised Vocabulary-Informed Learning (120 citations)
- Multi-Cue Zero-Shot Learning With Strong Supervision (108 citations)
- Recovering the Missing Link: Predicting Class-Attribute Associations for Unsupervised Zero-Shot Learning (86 citations)
- One-Shot Learning of Scene Locations via Feature Trajectory Transfer (51 citations)
Self-supervised and contrastive · 2 papers
- Homography Estimation From the Common Self-Polar Triangle of Separate Ellipses (15 citations)
- Identifying Good Training Data for Self-Supervised Free Space Estimation (7 citations)
Noisy labels · 2 papers
- Seeing Through the Human Reporting Bias: Visual Classifiers From Noisy Human-Centric Labels (215 citations)
- Noisy Label Recovery for Shadow Detection in Unfamiliar Domains (42 citations)
Semi-supervised learning · 1 paper
- SemiContour: A Semi-Supervised Learning Approach for Contour Detection (48 citations)
