Few Shot Image Classification On Tiered

평가 지표

Accuracy

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	Accuracy	Paper Title	Repository
TapNet	63.08	TapNet: Neural Network Augmented with Task-Adaptive Projection for Few-Shot Learning	-
DiffKendall (Meta-Baseline, ResNet-12)	70.76	DiffKendall: A Novel Approach for Few-Shot Learning with Differentiable Kendall's Rank Correlation	-
CAML [Laion-2b]	96.8	Context-Aware Meta-Learning	-
HCTransformers	79.67	Attribute Surrogates Learning and Spectral Tokens Pooling in Transformers for Few-shot Learning	-
Adaptive Subspace Network	68.44	Adaptive Subspaces for Few-Shot Learning
LST	77.7	Learning to Self-Train for Semi-Supervised Few-Shot Classification	-
BD-CSPN + ESFR (ResNet-18)	80.13	Unsupervised Embedding Adaptation via Early-Stage Feature Reconstruction for Few-Shot Classification	-
RENet	71.61	Relational Embedding for Few-Shot Classification	-
ASY ResNet12 (transductive)	83.98	EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients	-
MTUNet+WRN	62.42	Match Them Up: Visually Explainable Few-shot Image Classification	-
TIM-GD	82.1	Transductive Information Maximization For Few-Shot Learning	-
EASY 3xResNet12 (inductive)	74.71	EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients	-
GML (ResNet-12)	69.61	Geometric Mean Improves Loss For Few-Shot Learning	-
ASY ResNet12 (ours)	74.31	EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients	-
SImPa	70.82	PAC-Bayes meta-learning with implicit task-specific posteriors	-
Deep Comparison Network	68.83	RelationNet2: Deep Comparison Columns for Few-Shot Learning	-
AM3-TADAM	69.08	Adaptive Cross-Modal Few-Shot Learning	-
Transductive CNAPS	65.9	Enhancing Few-Shot Image Classification with Unlabelled Examples	-
Simple CNAPS + FETI	71.4	Improved Few-Shot Visual Classification	-
BAVARDAGE	85.20	Adaptive Dimension Reduction and Variational Inference for Transductive Few-Shot Classification	-

0 of 49 row(s) selected.