Few Shot Image Classification On Mini 2

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 Title	Repository
SgVA-CLIP	97.95	SgVA-CLIP: Semantic-guided Visual Adapting of Vision-Language Models for Few-shot Image Classification
CAML [Laion-2b]	96.2	Context-Aware Meta-Learning
P>M>F (P=DINO-ViT-base, M=ProtoNet)	95.3	Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and Fine-Tuning Make a Difference
TRIDENT	86.11	Transductive Decoupled Variational Inference for Few-Shot Classification
PT+MAP+SF+SOT (transductive)	85.59	The Self-Optimal-Transport Feature Transform
PT+MAP+SF+BPA (transductive)	85.59	The Balanced-Pairwise-Affinities Feature Transform
PEMnE-BMS* (transductive)	85.54	Squeezing Backbone Feature Distributions to the Max for Efficient Few-Shot Learning
PT+MAP (s+f) (transductive)	84.81	Few-Shot Learning by Integrating Spatial and Frequency Representation
BAVARDAGE	84.80	Adaptive Dimension Reduction and Variational Inference for Transductive Few-Shot Classification	-
EASY 3xResNet12 (transductive)	84.04	EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients
Illumination Augmentation	82.99	Sill-Net: Feature Augmentation with Separated Illumination Representation
PT+MAP (transductive)	82.92	Leveraging the Feature Distribution in Transfer-based Few-Shot Learning
EASY 2xResNet12 1/√2 (transductive)	82.31	EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients
Transductive CNAPS + FETI	79.9	Enhancing Few-Shot Image Classification with Unlabelled Examples
PrototypeCompletion	79.01	Prototype Completion for Few-Shot Learning
SemFew-Trans	78.94	Simple Semantic-Aided Few-Shot Learning
MCT	78.55	Meta-Learned Confidence for Few-shot Learning
TIM-GD	77.80	Transductive Information Maximization For Few-Shot Learning
Simple CNAPS + FETI	77.4	Improved Few-Shot Visual Classification
EPNet	77.27	Embedding Propagation: Smoother Manifold for Few-Shot Classification

0 of 105 row(s) selected.