SOTAVerified

Few-Shot Image Classification

Few-Shot Image Classification is a computer vision task that involves training machine learning models to classify images into predefined categories using only a few labeled examples of each category (typically ( Image credit: Learning Embedding Adaptation for Few-Shot Learning )

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 51–100 of 353 papers

TitleStatusHype
Learning to Compare: Relation Network for Few-Shot LearningCode1
Context-Aware Meta-LearningCode1
Contextual Squeeze-and-Excitation for Efficient Few-Shot Image ClassificationCode1
Prompt Tuning based Adapter for Vision-Language Model AdaptionCode1
BaseTransformers: Attention over base data-points for One Shot LearningCode1
Language Quantized AutoEncoders: Towards Unsupervised Text-Image AlignmentCode1
BECLR: Batch Enhanced Contrastive Few-Shot LearningCode1
Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and Fine-Tuning Make a DifferenceCode1
Bi-directional Feature Reconstruction Network for Fine-Grained Few-Shot Image ClassificationCode1
Language Models as Black-Box Optimizers for Vision-Language ModelsCode1
Bongard-HOI: Benchmarking Few-Shot Visual Reasoning for Human-Object InteractionsCode1
Debiased Learning from Naturally Imbalanced Pseudo-LabelsCode1
Leveraging the Feature Distribution in Transfer-based Few-Shot LearningCode1
Bayesian Meta-Learning for the Few-Shot Setting via Deep KernelsCode1
Few-Shot Image Classification Benchmarks are Too Far From Reality: Build Back Better with Semantic Task SamplingCode1
Laplacian Regularized Few-Shot LearningCode1
Few-Shot Learning by Integrating Spatial and Frequency RepresentationCode1
Few-Shot Learning via Embedding Adaptation with Set-to-Set FunctionsCode1
Bridging Multi-Task Learning and Meta-Learning: Towards Efficient Training and Effective AdaptationCode1
DeepEMD: Differentiable Earth Mover's Distance for Few-Shot LearningCode1
A Broader Study of Cross-Domain Few-Shot LearningCode1
Diffusion Mechanism in Residual Neural Network: Theory and ApplicationsCode1
Few-shot Image Classification: Just Use a Library of Pre-trained Feature Extractors and a Simple ClassifierCode1
Disentangled Feature Representation for Few-shot Image ClassificationCode1
Distilling Large Vision-Language Model with Out-of-Distribution GeneralizabilityCode1
Charting the Right Manifold: Manifold Mixup for Few-shot LearningCode1
Class-Aware Patch Embedding Adaptation for Few-Shot Image ClassificationCode1
Diversified in-domain synthesis with efficient fine-tuning for few-shot classificationCode1
Meta-Album: Multi-domain Meta-Dataset for Few-Shot Image ClassificationCode1
Dynamic Few-Shot Visual Learning without ForgettingCode1
EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple IngredientsCode1
GPU-based Self-Organizing Maps for Post-Labeled Few-Shot Unsupervised LearningCode1
Generalized Few-Shot Video Classification with Video Retrieval and Feature GenerationCode1
Large Language Models are Good Prompt Learners for Low-Shot Image ClassificationCode1
Embedding Propagation: Smoother Manifold for Few-Shot ClassificationCode1
Empirical Bayes Transductive Meta-Learning with Synthetic GradientsCode1
Enhancing Few-Shot Image Classification through Learnable Multi-Scale Embedding and Attention MechanismsCode1
Enhancing Few-shot Image Classification with Cosine TransformerCode1
How to train your MAMLCode1
Meta-Transfer Learning for Few-Shot LearningCode1
ePillID Dataset: A Low-Shot Fine-Grained Benchmark for Pill IdentificationCode1
ESPT: A Self-Supervised Episodic Spatial Pretext Task for Improving Few-Shot LearningCode1
HyperTransformer: Model Generation for Supervised and Semi-Supervised Few-Shot LearningCode1
Negative Margin Matters: Understanding Margin in Few-shot ClassificationCode1
Memory Efficient Meta-Learning with Large ImagesCode1
Improving ProtoNet for Few-Shot Video Object Recognition: Winner of ORBIT Challenge 2022Code1
Adaptive Subspaces for Few-Shot LearningCode1
Exploring Complementary Strengths of Invariant and Equivariant Representations for Few-Shot LearningCode1
Fast and Flexible Multi-Task Classification Using Conditional Neural Adaptive ProcessesCode1
ORBIT: A Real-World Few-Shot Dataset for Teachable Object RecognitionCode1
Show:102550
← PrevPage 2 of 8Next →
All datasetsMini-Imagenet 5-way (1-shot)Mini-Imagenet 5-way (5-shot)Tiered ImageNet 5-way (5-shot)Tiered ImageNet 5-way (1-shot)CIFAR-FS 5-way (5-shot)CIFAR-FS 5-way (1-shot)CUB 200 5-way 1-shotCUB 200 5-way 5-shotFC100 5-way (1-shot)FC100 5-way (5-shot)Meta-DatasetOMNIGLOT - 1-Shot, 20-way

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1SgVA-CLIPAccuracy97.95—Unverified
2CAML [Laion-2b]Accuracy96.2—Unverified
3P>M>F (P=DINO-ViT-base, M=ProtoNet)Accuracy95.3—Unverified
4TRIDENTAccuracy86.11—Unverified
5PT+MAP+SF+SOT (transductive)Accuracy85.59—Unverified
6PT+MAP+SF+BPA (transductive)Accuracy85.59—Unverified
7PEMnE-BMS* (transductive)Accuracy85.54—Unverified
8PT+MAP (s+f) (transductive)Accuracy84.81—Unverified
9BAVARDAGEAccuracy84.8—Unverified
10EASY 3xResNet12 (transductive)Accuracy84.04—Unverified
#ModelMetricClaimedVerifiedStatus
1SgVA-CLIPAccuracy98.72—Unverified
2CAML [Laion-2b]Accuracy98.6—Unverified
3P>M>F (P=DINO-ViT-base, M=ProtoNet)Accuracy98.4—Unverified
4TRIDENTAccuracy95.95—Unverified
5BAVARDAGEAccuracy91.65—Unverified
6PEMnE-BMS*(transductive)Accuracy91.53—Unverified
7Transductive CNAPS + FETIAccuracy91.5—Unverified
8PT+MAP+SF+BPA (transductive)Accuracy91.34—Unverified
9PT+MAP+SF+SOT (transductive)Accuracy91.34—Unverified
10AmdimNetAccuracy90.98—Unverified