ECCV 2024 accepted papers on Label-efficient and robust learning
247 of the 2,387 papers accepted at ECCV 2024 (10.3%) are on Label-efficient and robust learning, compared with 16.3% at ECCV 2022. They fall into 9 topics.
See ECCV 2024 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 · 71 papers
- Attention-Challenging Multiple Instance Learning for Whole Slide Image Classification (111 citations)
- Multi-Memory Matching for Unsupervised Visible-Infrared Person Re-Identification (72 citations)
- Leveraging Enhanced Queries of Point Sets for Vectorized Map Construction (57 citations)
- Discover-then-Name: Task-Agnostic Concept Bottlenecks via Automated Concept Discovery (41 citations)
- DGR-MIL: Exploring Diverse Global Representation in Multiple Instance Learning for Whole Slide Image Classification (39 citations)
- Keypoint Promptable Re-Identification (37 citations)
- PetFace: A Large-Scale Dataset and Benchmark for Animal Identification (29 citations)
- PPAD: Iterative Interactions of Prediction and Planning for End-to-end Autonomous Driving (28 citations)
- Causality-inspired Discriminative Feature Learning in Triple Domains for Gait Recognition (26 citations)
- NeuroNCAP: Photorealistic Closed-loop Safety Testing for Autonomous Driving (26 citations)
- ADMap: Anti-disturbance Framework for Vectorized HD Map Construction (25 citations)
- Enhancing Vectorized Map Perception with Historical Rasterized Maps (25 citations)
and 59 more in the interactive map.
Domain adaptation and generalization · 36 papers
- FairDomain: Achieving Fairness in Cross-Domain Medical Image Segmentation and Classification (29 citations)
- Generalizable Facial Expression Recognition (29 citations)
- Bottom-Up Domain Prompt Tuning for Generalized Face Anti-Spoofing (19 citations)
- Multimodal Cross-Domain Few-Shot Learning for Egocentric Action Recognition (19 citations)
- CoDA: Instructive Chain-of-Domain Adaptation with Severity-Aware Visual Prompt Tuning (17 citations)
- MICDrop: Masking Image and Depth Features via Complementary Dropout for Domain-Adaptive Semantic Segmentation (17 citations)
- Towards Multimodal Open-Set Domain Generalization and Adaptation through Self-supervision (15 citations)
- Training-Free Model Merging for Multi-target Domain Adaptation (15 citations)
- Rethinking Normalization Layers for Domain Generalizable Person Re-identification (13 citations)
- De-Confusing Pseudo-Labels in Source-Free Domain Adaptation (12 citations)
- COD: Learning Conditional Invariant Representation for Domain Adaptation Regression (9 citations)
- Cross-Domain Semantic Segmentation on Inconsistent Taxonomy using VLMs (8 citations)
and 24 more in the interactive map.
Self-supervised and contrastive · 34 papers
- OmniSat: Self-Supervised Modality Fusion for Earth Observation (61 citations)
- Multistain Pretraining for Slide Representation Learning in Pathology (38 citations)
- Masked Angle-Aware Autoencoder for Remote Sensing Images (32 citations)
- InfMAE: A Foundation Model in The Infrared Modality (28 citations)
- SSL-Cleanse: Trojan Detection and Mitigation in Self-Supervised Learning (26 citations)
- Region-centric Image-Language Pretraining for Open-Vocabulary Detection (18 citations)
- Towards Latent Masked Image Modeling for Self-Supervised Visual Representation Learning (17 citations)
- HYPE: Hyperbolic Entailment Filtering for Underspecified Images and Texts (16 citations)
- Learning Representations of Satellite Images From Metadata Supervision (13 citations)
- Snuffy: Efficient Whole Slide Image Classifier (13 citations)
- On Pretraining Data Diversity for Self-Supervised Learning (12 citations)
- S-JEPA: A Joint Embedding Predictive Architecture for Skeletal Action Recognition (12 citations)
and 22 more in the interactive map.
OOD and anomaly detection · 28 papers
- GLAD: Towards Better Reconstruction with Global and Local Adaptive Diffusion Models for Unsupervised Anomaly Detection (91 citations)
- GalLop: Learning global and local prompts for vision-language models (42 citations)
- GeneralAD: Anomaly Detection Across Domains by Attending to Distorted Features (37 citations)
- Self-supervised Feature Adaptation for 3D Industrial Anomaly Detection (33 citations)
- Learning Unified Reference Representation for Unsupervised Multi-class Anomaly Detection (32 citations)
- Hierarchical Gaussian Mixture Normalizing Flow Modeling for Unified Anomaly Detection (31 citations)
- Few-Shot Anomaly-Driven Generation for Anomaly Classification and Segmentation (29 citations)
- Continuous Memory Representation for Anomaly Detection (26 citations)
- MoEAD: A Parameter-efficient Model for Multi-class Anomaly Detection (19 citations)
- LAPT: Label-driven Automated Prompt Tuning for OOD Detection with Vision-Language Models (17 citations)
- Learning to Detect Multi-class Anomalies with Just One Normal Image Prompt (16 citations)
- Can OOD Object Detectors Learn from Foundation Models? (14 citations)
and 16 more in the interactive map.
Knowledge distillation · 23 papers
- Dataset Distillation by Automatic Training Trajectories (28 citations)
- UNIC: Universal Classification Models via Multi-teacher Distillation (23 citations)
- Improving Knowledge Distillation via Regularizing Feature Direction and Norm (22 citations)
- Good Teachers Explain: Explanation-Enhanced Knowledge Distillation (21 citations)
- AMD: Automatic Multi-step Distillation of Large-scale Vision Models (19 citations)
- Is Retain Set All You Need in Machine Unlearning? Restoring Performance of Unlearned Models with Out-Of-Distribution Images (17 citations)
- AdaDistill: Adaptive Knowledge Distillation for Deep Face Recognition (15 citations)
- Direct Distillation between Different Domains (15 citations)
- BKDSNN: Enhancing the Performance of Learning-based Spiking Neural Networks Training with Blurred Knowledge Distillation (13 citations)
- How to Train the Teacher Model for Effective Knowledge Distillation (13 citations)
- Distill Gold from Massive Ores: Bi-level Data Pruning towards Efficient Dataset Distillation (9 citations)
- Make a Strong Teacher with Label Assistance: A Novel Knowledge Distillation Approach for Semantic Segmentation (8 citations)
and 11 more in the interactive map.
Semi-supervised learning · 21 papers
- Alternate Diverse Teaching for Semi-supervised Medical Image Segmentation (65 citations)
- SemiVL: Semi-Supervised Semantic Segmentation with Vision-Language Guidance (55 citations)
- Beyond Pixels: Semi-Supervised Semantic Segmentation with a Multi-scale Patch-based Multi-Label Classifier (26 citations)
- LMT-GP: Combined Latent Mean-Teacher and Gaussian Process for Semi-supervised Low-light Image Enhancement (14 citations)
- PMT: Progressive Mean Teacher via Exploring Temporal Consistency for Semi-Supervised Medical Image Segmentation (13 citations)
- Semi-Supervised Teacher-Reference-Student Architecture for Action Quality Assessment (12 citations)
- Semi-supervised Segmentation of Histopathology Images with Noise-Aware Topological Consistency (11 citations)
- Gradient-Aware for Class-Imbalanced Semi-supervised Medical Image Segmentation (8 citations)
- SCOMatch: Alleviating Overtrusting in Open-set Semi-supervised Learning (6 citations)
- The Devil is in the Statistics: Mitigating and Exploiting Statistics Difference for Generalizable Semi-supervised Medical Image Segmentation (6 citations)
- Towards Adaptive Pseudo-label Learning for Semi-Supervised Temporal Action Localization (6 citations)
- ItTakesTwo: Leveraging Peer Representations for Semi-supervised LiDAR Semantic Segmentation (5 citations)
and 9 more in the interactive map.
Test-time adaptation · 14 papers
- Prompt-Based Test-Time Real Image Dehazing: A Novel Pipeline (25 citations)
- TTT-MIM: Test-Time Training with Masked Image Modeling for Denoising Distribution Shifts (10 citations)
- STAMP: Outlier-Aware Test-Time Adaptation with Stable Memory Replay (9 citations)
- Distribution Alignment for Fully Test-Time Adaptation with Dynamic Online Data Streams (8 citations)
- Efficient Diffusion-Driven Corruption Editor for Test-Time Adaptation (8 citations)
- MemBN: Robust Test-Time Adaptation via Batch Norm with Statistics Memory (8 citations)
- Bi-TTA: Bidirectional Test-Time Adapter for Remote Physiological Measurement (7 citations)
- Reliable Spatial-Temporal Voxels For Multi-Modal Test-Time Adaptation (7 citations)
- UL-VIO: Ultra-lightweight Visual-Inertial Odometry with Noise Robust Test-time Adaptation (7 citations)
- Domain-adaptive Video Deblurring via Test-time Blurring (6 citations)
- Dual-Path Adversarial Lifting for Domain Shift Correction in Online Test-time Adaptation (5 citations)
- Test-Time Stain Adaptation with Diffusion Models for Histopathology Image Classification (4 citations)
and 2 more in the interactive map.
Zero-/few-shot and meta-learning · 11 papers
- SA-DVAE: Improving Zero-Shot Skeleton-Based Action Recognition by Disentangled Variational Autoencoders (21 citations)
- Improving Zero-shot Generalization of Learned Prompts via Unsupervised Knowledge Distillation (16 citations)
- Learning Representations from Foundation Models for Domain Generalized Stereo Matching (15 citations)
- C2C: Component-to-Composition Learning for Zero-Shot Compositional Action Recognition (11 citations)
- Zero-Shot Image Feature Consensus with Deep Functional Maps (8 citations)
- Improving Zero-Shot Generalization for CLIP with Variational Adapter (7 citations)
- MRSP: Learn Multi-Representations of Single Primitive for Compositional Zero-Shot Learning (5 citations)
- Do Generalised Classifiers really work on Human Drawn Sketches? (2 citations)
- Learning to Obstruct Few-Shot Image Classification over Restricted Classes (2 citations)
- Benchmarking Spurious Bias in Few-Shot Image Classifiers (1 citation)
- Exploring Active Learning in Meta-Learning: Enhancing Context Set Labeling (1 citation)
Noisy labels · 9 papers
- Foster Adaptivity and Balance in Learning with Noisy Labels (16 citations)
- Instance-dependent Noisy-label Learning with Graphical Model Based Noise-rate Estimation (6 citations)
- Distribution-Aware Robust Learning from Long-Tailed Data with Noisy Labels (4 citations)
- Understanding and Mitigating Human-Labelling Errors in Supervised Contrastive Learning (4 citations)
- MTaDCS: Moving Trace and Feature Density-based Confidence Sample Selection under Label Noise (3 citations)
- LNL+K: Enhancing Learning with Noisy Labels Through Noise Source Knowledge Integration (2 citations)
- Optimal Transport of Diverse Unsupervised Tasks for Robust Learning from Noisy Few-Shot Data (2 citations)
- An accurate detection is not all you need to combat label noise in web-noisy datasets (1 citation)
- Norma: A Noise Robust Memory-Augmented Framework for Whole Slide Image Classification (1 citation)
