Few Shot Image Classification On Cifar Fs 5

평가 지표

Accuracy

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	Accuracy	Paper Title	Repository
MCRNet-RR	73.8	Complementing Representation Deficiency in Few-shot Image Classification: A Meta-Learning Approach
MetaOptNet-SVM+Task Aug	76.75	Task Augmentation by Rotating for Meta-Learning
Illumination Augmentation	87.73	Sill-Net: Feature Augmentation with Separated Illumination Representation
PT+MAP+SF+SOT (transductive)	89.94	The Self-Optimal-Transport Feature Transform
MetaQDA	75.83	Shallow Bayesian Meta Learning for Real-World Few-Shot Recognition
BAVARDAGE	87.35	Adaptive Dimension Reduction and Variational Inference for Transductive Few-Shot Classification	-
MTUNet+ResNet-18	66.31	Match Them Up: Visually Explainable Few-shot Image Classification
SIB	80.0	Empirical Bayes Transductive Meta-Learning with Synthetic Gradients
ConstellationNets	75.4	Constellation Nets for Few-Shot Learning
ACC + Amphibian	73.1	Generalized Adaptation for Few-Shot Learning	-
Invariance-Equivariance	77.87	Exploring Complementary Strengths of Invariant and Equivariant Representations for Few-Shot Learning
RENet	74.51	Relational Embedding for Few-Shot Classification
ICI	76.51	Instance Credibility Inference for Few-Shot Learning
GML (ResNet-12)	71.09	Geometric Mean Improves Loss For Few-Shot Learning	-
EASY 2xResNet12 1/√2 (transductive)	86.99	EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients
EASY 3xResNet12 (transductive)	87.16	EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients
pseudo-shots	81.87	Extended Few-Shot Learning: Exploiting Existing Resources for Novel Tasks
Adaptive Subspace Network	78	Adaptive Subspaces for Few-Shot Learning
EASY 2xResNet12 1/√2 (inductive)	75.24	EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients
PT+MAP+SF+BPA (transductive)	89.94	The Balanced-Pairwise-Affinities Feature Transform

0 of 38 row(s) selected.