Few Shot Image Classification On Tiered

评估指标

Accuracy

评测结果

各个模型在此基准测试上的表现结果

Paper TitleRepository
CAML [Laion-2b]96.8Context-Aware Meta-Learning
TRIDENT86.97Transductive Decoupled Variational Inference for Few-Shot Classification
PEMnE-BMS*86.07Squeezing Backbone Feature Distributions to the Max for Efficient Few-Shot Learning
PT+MAP85.41Leveraging the Feature Distribution in Transfer-based Few-Shot Learning
BAVARDAGE85.20Adaptive Dimension Reduction and Variational Inference for Transductive Few-Shot Classification-
EASY 3xResNet12 (transductive)84.29EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients
ICI84.01Instance Credibility Inference for Few-Shot Learning
ASY ResNet12 (transductive)83.98EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients
TIM-GD82.1Transductive Information Maximization For Few-Shot Learning
BD-CSPN + ESFR (WRN)81.77Unsupervised Embedding Adaptation via Early-Stage Feature Reconstruction for Few-Shot Classification
LaplacianShot80.30Laplacian Regularized Few-Shot Learning-
BD-CSPN + ESFR (ResNet-18)80.13Unsupervised Embedding Adaptation via Early-Stage Feature Reconstruction for Few-Shot Classification
HCTransformers79.67Attribute Surrogates Learning and Spectral Tokens Pooling in Transformers for Few-shot Learning
EPNet78.50Embedding Propagation: Smoother Manifold for Few-Shot Classification
LST77.7Learning to Self-Train for Semi-Supervised Few-Shot Classification
OSLO76.64Open-Set Likelihood Maximization for Few-Shot Learning
pseudo-shots76.55Extended Few-Shot Learning: Exploiting Existing Resources for Novel Tasks
FewTURE76.32Rethinking Generalization in Few-Shot Classification
EASY 3xResNet12 (inductive)74.71EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients
MetaQDA74.33Shallow Bayesian Meta Learning for Real-World Few-Shot Recognition
0 of 49 row(s) selected.
Few Shot Image Classification On Tiered | SOTA | HyperAI超神经