Few Shot Image Classification On Cub 200 5

Métriques

Accuracy

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	Accuracy	Paper Title	Repository
CAML [Laion-2b]	98.7	Context-Aware Meta-Learning	-
LaplacianShot	88.68	Laplacian Regularized Few-Shot Learning
High-End MAML++	83.8	Learning to learn via Self-Critique
S2M2R	90.85	Charting the Right Manifold: Manifold Mixup for Few-shot Learning
EASY 4xResNet12 (inductive)	91.59	EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients
DN4-DA (k=1)	81.9	Revisiting Local Descriptor based Image-to-Class Measure for Few-shot Learning
DKT + BNCosSim	85.64	Bayesian Meta-Learning for the Few-Shot Setting via Deep Kernels
PT+MAP+SF+BPA (transductive)	97.12	The Balanced-Pairwise-Affinities Feature Transform
Hyperbolic ProtoNet	72.22	Hyperbolic Image Embeddings
ICI	92.48	Instance Credibility Inference for Few-Shot Learning
EASY 4xResNet12 (transductive)	93.5	EASY: Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients
BAVARDAGE	93.50	Adaptive Dimension Reduction and Variational Inference for Transductive Few-Shot Classification	-
TDM	93.37	Task Discrepancy Maximization for Fine-grained Few-Shot Classification
PT+MAP	93.99	Leveraging the Feature Distribution in Transfer-based Few-Shot Learning
TIM-GD	90.8	Transductive Information Maximization For Few-Shot Learning
Illumination Augmentation	96.28	Sill-Net: Feature Augmentation with Separated Illumination Representation
VFD	91.48	Variational Feature Disentangling for Fine-Grained Few-Shot Classification	-
ESPT	94.02	ESPT: A Self-Supervised Episodic Spatial Pretext Task for Improving Few-Shot Learning
BD-CSPN + ESFR (ResNet-18)	88.65	Unsupervised Embedding Adaptation via Early-Stage Feature Reconstruction for Few-Shot Classification
LST+MAP	94.09	Transfer learning based few-shot classification using optimal transport mapping from preprocessed latent space of backbone neural network

0 of 32 row(s) selected.