CVPR 2023 accepted papers on Label-efficient and robust learning

Recognition and long-tail · 111 papers

• Curricular Contrastive Regularization for Physics-Aware Single Image Dehazing (336 citations)
• Diverse Embedding Expansion Network and Low-Light Cross-Modality Benchmark for Visible-Infrared Person Re-Identification (317 citations)
• V2X-Seq: A Large-Scale Sequential Dataset for Vehicle-Infrastructure Cooperative Perception and Forecasting (176 citations)
• Implicit Identity Driven Deepfake Face Swapping Detection (168 citations)
• Shape-Erased Feature Learning for Visible-Infrared Person Re-Identification (159 citations)
• Language in a Bottle: Language Model Guided Concept Bottlenecks for Interpretable Image Classification (151 citations)
• Interventional Bag Multi-Instance Learning on Whole-Slide Pathological Images (138 citations)
• Good Is Bad: Causality Inspired Cloth-Debiasing for Cloth-Changing Person Re-Identification (133 citations)
• Micron-BERT: BERT-Based Facial Micro-Expression Recognition (133 citations)
• PartMix: Regularization Strategy To Learn Part Discovery for Visible-Infrared Person Re-Identification (132 citations)
• PyramidFlow: High-Resolution Defect Contrastive Localization Using Pyramid Normalizing Flow (129 citations)
• Think Twice Before Driving: Towards Scalable Decoders for End-to-End Autonomous Driving (111 citations)

and 99 more in the interactive map.

Domain adaptation and generalization · 56 papers

• MIC: Masked Image Consistency for Context-Enhanced Domain Adaptation (350 citations)
• Guiding Pseudo-Labels With Uncertainty Estimation for Source-Free Unsupervised Domain Adaptation (138 citations)
• Contrastive Mean Teacher for Domain Adaptive Object Detectors (137 citations)
• Sharpness-Aware Gradient Matching for Domain Generalization (118 citations)
• Instance-Aware Domain Generalization for Face Anti-Spoofing (101 citations)
• Rethinking Domain Generalization for Face Anti-Spoofing: Separability and Alignment (100 citations)
• Patch-Mix Transformer for Unsupervised Domain Adaptation: A Game Perspective (99 citations)
• Deep Frequency Filtering for Domain Generalization (93 citations)
• StyleAdv: Meta Style Adversarial Training for Cross-Domain Few-Shot Learning (91 citations)
• C-SFDA: A Curriculum Learning Aided Self-Training Framework for Efficient Source Free Domain Adaptation (89 citations)
• ALOFT: A Lightweight MLP-Like Architecture With Dynamic Low-Frequency Transform for Domain Generalization (81 citations)
• Harmonious Teacher for Cross-Domain Object Detection (70 citations)

and 44 more in the interactive map.

Self-supervised and contrastive · 52 papers

• Self-Supervised Learning From Images With a Joint-Embedding Predictive Architecture (399 citations)
• Benchmarking Self-Supervised Learning on Diverse Pathology Datasets (179 citations)
• Beyond Appearance: A Semantic Controllable Self-Supervised Learning Framework for Human-Centric Visual Tasks (154 citations)
• Change-Aware Sampling and Contrastive Learning for Satellite Images (94 citations)
• Actionlet-Dependent Contrastive Learning for Unsupervised Skeleton-Based Action Recognition (91 citations)
• Siamese Image Modeling for Self-Supervised Vision Representation Learning (87 citations)
• Best of Both Worlds: Multimodal Contrastive Learning With Tabular and Imaging Data (70 citations)
• LG-BPN: Local and Global Blind-Patch Network for Self-Supervised Real-World Denoising (69 citations)
• SelfME: Self-Supervised Motion Learning for Micro-Expression Recognition (59 citations)
• On the Effects of Self-Supervision and Contrastive Alignment in Deep Multi-View Clustering (57 citations)
• On Data Scaling in Masked Image Modeling (56 citations)
• Learning Audio-Visual Source Localization via False Negative Aware Contrastive Learning (52 citations)

and 40 more in the interactive map.

Semi-supervised learning · 31 papers

• Bidirectional Copy-Paste for Semi-Supervised Medical Image Segmentation (412 citations)
• Contrastive Semi-Supervised Learning for Underwater Image Restoration via Reliable Bank (269 citations)
• Pseudo-Label Guided Contrastive Learning for Semi-Supervised Medical Image Segmentation (200 citations)
• MCF: Mutual Correction Framework for Semi-Supervised Medical Image Segmentation (199 citations)
• Augmentation Matters: A Simple-Yet-Effective Approach to Semi-Supervised Semantic Segmentation (147 citations)
• Conflict-Based Cross-View Consistency for Semi-Supervised Semantic Segmentation (97 citations)
• PromptCAL: Contrastive Affinity Learning via Auxiliary Prompts for Generalized Novel Category Discovery (79 citations)
• Boosting Semi-Supervised Learning by Exploiting All Unlabeled Data (63 citations)
• PEFAT: Boosting Semi-Supervised Medical Image Classification via Pseudo-Loss Estimation and Feature Adversarial Training (59 citations)
• Instance-Specific and Model-Adaptive Supervision for Semi-Supervised Semantic Segmentation (56 citations)
• Hunting Sparsity: Density-Guided Contrastive Learning for Semi-Supervised Semantic Segmentation (55 citations)
• MagicNet: Semi-Supervised Multi-Organ Segmentation via Magic-Cube Partition and Recovery (55 citations)

and 19 more in the interactive map.

Zero-/few-shot and meta-learning · 26 papers

• Pic2Word: Mapping Pictures to Words for Zero-Shot Composed Image Retrieval (135 citations)
• Revisiting Prototypical Network for Cross Domain Few-Shot Learning (98 citations)
• Finetune Like You Pretrain: Improved Finetuning of Zero-Shot Vision Models (90 citations)
• I2MVFormer: Large Language Model Generated Multi-View Document Supervision for Zero-Shot Image Classification (81 citations)
• Transductive Few-Shot Learning With Prototype-Based Label Propagation by Iterative Graph Refinement (77 citations)
• Progressive Semantic-Visual Mutual Adaption for Generalized Zero-Shot Learning (72 citations)
• Diversity-Aware Meta Visual Prompting (60 citations)
• Learning Conditional Attributes for Compositional Zero-Shot Learning (57 citations)
• Glocal Energy-Based Learning for Few-Shot Open-Set Recognition (44 citations)
• Prompt-Guided Zero-Shot Anomaly Action Recognition Using Pretrained Deep Skeleton Features (43 citations)
• Bi-Level Meta-Learning for Few-Shot Domain Generalization (36 citations)
• ProD: Prompting-To-Disentangle Domain Knowledge for Cross-Domain Few-Shot Image Classification (36 citations)

and 14 more in the interactive map.

OOD and anomaly detection · 18 papers

• SimpleNet: A Simple Network for Image Anomaly Detection and Localization (619 citations)
• Implicit Identity Leakage: The Stumbling Block to Improving Deepfake Detection Generalization (202 citations)
• Unbiased Multiple Instance Learning for Weakly Supervised Video Anomaly Detection (168 citations)
• Exploiting Completeness and Uncertainty of Pseudo Labels for Weakly Supervised Video Anomaly Detection (142 citations)
• Explicit Boundary Guided Semi-Push-Pull Contrastive Learning for Supervised Anomaly Detection (108 citations)
• Diversity-Measurable Anomaly Detection (107 citations)
• GEN: Pushing the Limits of Softmax-Based Out-of-Distribution Detection (85 citations)
• Rethinking Out-of-Distribution (OOD) Detection: Masked Image Modeling Is All You Need (56 citations)
• Decoupling MaxLogit for Out-of-Distribution Detection (47 citations)
• Block Selection Method for Using Feature Norm in Out-of-Distribution Detection (36 citations)
• OpenMix: Exploring Outlier Samples for Misclassification Detection (32 citations)
• Balanced Energy Regularization Loss for Out-of-Distribution Detection (31 citations)

and 6 more in the interactive map.

Knowledge distillation · 16 papers

• Class Attention Transfer Based Knowledge Distillation (155 citations)
• Multi-Level Logit Distillation (139 citations)
• MMANet: Margin-Aware Distillation and Modality-Aware Regularization for Incomplete Multimodal Learning (67 citations)
• DisWOT: Student Architecture Search for Distillation WithOut Training (65 citations)
• Masked Autoencoders Enable Efficient Knowledge Distillers (57 citations)
• Rethinking Feature-Based Knowledge Distillation for Face Recognition (48 citations)
• Data-Free Knowledge Distillation via Feature Exchange and Activation Region Constraint (44 citations)
• Accelerating Dataset Distillation via Model Augmentation (43 citations)
• Multi-Mode Online Knowledge Distillation for Self-Supervised Visual Representation Learning (40 citations)
• DistilPose: Tokenized Pose Regression With Heatmap Distillation (37 citations)
• Learning To Retain While Acquiring: Combating Distribution-Shift in Adversarial Data-Free Knowledge Distillation (33 citations)
• StructVPR: Distill Structural Knowledge With Weighting Samples for Visual Place Recognition (28 citations)

and 4 more in the interactive map.

Test-time adaptation · 15 papers

• Robust Test-Time Adaptation in Dynamic Scenarios (117 citations)
• Robust Mean Teacher for Continual and Gradual Test-Time Adaptation (92 citations)
• Back to the Source: Diffusion-Driven Adaptation To Test-Time Corruption (70 citations)
• Feature Alignment and Uniformity for Test Time Adaptation (45 citations)
• Improved Test-Time Adaptation for Domain Generalization (44 citations)
• TTA-COPE: Test-Time Adaptation for Category-Level Object Pose Estimation (38 citations)
• Dynamically Instance-Guided Adaptation: A Backward-Free Approach for Test-Time Domain Adaptive Semantic Segmentation (36 citations)
• A Probabilistic Framework for Lifelong Test-Time Adaptation (31 citations)
• TeSLA: Test-Time Self-Learning With Automatic Adversarial Augmentation (31 citations)
• TIPI: Test Time Adaptation With Transformation Invariance (27 citations)
• Cloud-Device Collaborative Adaptation to Continual Changing Environments in the Real-World (20 citations)
• ActMAD: Activation Matching To Align Distributions for Test-Time-Training (18 citations)

and 3 more in the interactive map.

Noisy labels · 11 papers

• DISC: Learning From Noisy Labels via Dynamic Instance-Specific Selection and Correction (121 citations)
• Twin Contrastive Learning With Noisy Labels (115 citations)
• Fine-Grained Classification With Noisy Labels (56 citations)
• Noisy Correspondence Learning With Meta Similarity Correction (47 citations)
• Learning From Noisy Labels With Decoupled Meta Label Purifier (46 citations)
• RONO: Robust Discriminative Learning With Noisy Labels for 2D-3D Cross-Modal Retrieval (37 citations)
• OT-Filter: An Optimal Transport Filter for Learning With Noisy Labels (34 citations)
• Learning With Noisy Labels via Self-Supervised Adversarial Noisy Masking (25 citations)
• Leveraging Inter-Rater Agreement for Classification in the Presence of Noisy Labels (17 citations)
• Collaborative Noisy Label Cleaner: Learning Scene-Aware Trailers for Multi-Modal Highlight Detection in Movies (10 citations)
• How To Prevent the Continuous Damage of Noises To Model Training? (5 citations)