ML Research Wiki / Benchmarks / Few-Shot Image Classification / Meta-Dataset

Meta-Dataset

Few-Shot Image Classification Benchmark

Performance Over Time

📊 Showing 22 results | 📏 Metric: Accuracy

Top Performing Models

Rank	Model	Paper	Accuracy	Date	Code
1	SMAT (DINO-VIT-Base-16-224)	Unleashing the Power of Meta-tuning for Few-shot Generalization Through Sparse Interpolated Experts	85.27	2024-03-13	📦 szc12153/sparse_meta_tuning
2	P>M>F (P=DINO-ViT-base, M=ProtoNet)	Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and Fine-Tuning Make a Difference	84.75	2022-04-15	📦 hushell/pmf_cvpr22
3	TSP (ResNet18; applied on TA^2-Net)	Task-Specific Preconditioner for Cross-Domain Few-Shot Learning	81.40	2024-12-20	-
4	TSA (ResNet18, URL, residual adapters, 84x84 image, shuffled data, scratch, MDL)	Cross-domain Few-shot Learning with Task-specific Adapters	78.07	2021-07-01	📦 google-research/meta-dataset 📦 VICO-UoE/URL 📦 nobody-1617/deta 📦 jimzai/deta
5	UpperCaSE-EfficientNetB0	Contextual Squeeze-and-Excitation for Efficient Few-Shot Image Classification	76.10	2022-06-20	📦 mpatacchiola/contextual-squeeze-and-excitation
6	URL (ResNet18, 84x84 image, shuffled data, scratch, MDL)	Universal Representation Learning from Multiple Domains for Few-shot Classification	75.75	2021-03-25	📦 google-research/meta-dataset 📦 VICO-UoE/URL 📦 vico-uoe/universalrepresentations 📦 tmlr-group/CoPA 📦 tmlr-group/mokd
7	UpperCaSE-ResNet50	Contextual Squeeze-and-Excitation for Efficient Few-Shot Image Classification	74.90	2022-06-20	📦 mpatacchiola/contextual-squeeze-and-excitation
8	URT+MQDA	Shallow Bayesian Meta Learning for Real-World Few-Shot Recognition	74.30	2021-01-08	📦 open-debin/bayesian_mqda 📦 Open-Debin/MetaQDA_Pub
9	URT	A Universal Representation Transformer Layer for Few-Shot Image Classification	72.15	2020-06-21	📦 liulu112601/URT
10	SUR	Selecting Relevant Features from a Multi-domain Representation for Few-shot Classification	70.72	2020-03-20	📦 dvornikita/SUR

All Papers (22)

Unleashing the Power of Meta-tuning for Few-shot Generalization Through Sparse Interpolated Experts

2024

SMAT (DINO-VIT-Base-16-224)

szc12153/sparse_meta_tuning

Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and Fine-Tuning Make a Difference

2022

P>M>F (P=DINO-ViT-base, M=ProtoNet)

hushell/pmf_cvpr22

Task-Specific Preconditioner for Cross-Domain Few-Shot Learning

2024

TSP (ResNet18; applied on TA^2-Net)

Cross-domain Few-shot Learning with Task-specific Adapters

2021

TSA (ResNet18, URL, residual adapters, 84x84 image, shuffled data, scratch, MDL)

google-research/meta-dataset VICO-UoE/URL

Contextual Squeeze-and-Excitation for Efficient Few-Shot Image Classification

2022

UpperCaSE-EfficientNetB0

mpatacchiola/contextual-squeeze-and-excitation

Universal Representation Learning from Multiple Domains for Few-shot Classification

2021

URL (ResNet18, 84x84 image, shuffled data, scratch, MDL)

google-research/meta-dataset VICO-UoE/URL

Contextual Squeeze-and-Excitation for Efficient Few-Shot Image Classification

2022

UpperCaSE-ResNet50

mpatacchiola/contextual-squeeze-and-excitation

Shallow Bayesian Meta Learning for Real-World Few-Shot Recognition

2021

URT+MQDA

open-debin/bayesian_mqda Open-Debin/MetaQDA_Pub

A Universal Representation Transformer Layer for Few-Shot Image Classification

2020

URT

liulu112601/URT

Selecting Relevant Features from a Multi-domain Representation for Few-shot Classification

2020

SUR

dvornikita/SUR

Enhancing Few-Shot Image Classification with Unlabelled Examples

2020

Transductive CNAPS

peymanbateni/simple-cnaps plai-group/simple-cnaps

Improved Few-Shot Visual Classification

2019

Simple CNAPS

peymanbateni/simple-cnaps plai-group/simple-cnaps

Selecting Relevant Features from a Multi-domain Representation for Few-shot Classification

2020

SUR-pnf

dvornikita/SUR

Exploring Complementary Strengths of Invariant and Equivariant Representations for Few-Shot Learning

2021

Invariance-Equivariance

nayeemrizve/invariance-equivariance

Fast and Flexible Multi-Task Classification Using Conditional Neural Adaptive Processes

2019

CNAPs

cambridge-mlg/cnaps

Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples

2019

fo-Proto-MAML

google-research/meta-dataset cambridge-mlg/cnaps

Prototypical Networks for Few-shot Learning

2017

Prototypical Networks

learnables/learn2learn oscarknagg/few-shot

Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples

2019

Finetune

google-research/meta-dataset cambridge-mlg/cnaps

Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks

2017

fo-MAML

ray-project/ray PaddlePaddle/PaddleRec

Matching Networks for One Shot Learning

2016

Matching Networks

oscarknagg/few-shot sicara/easy-few-shot-learning

Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples

2019

k-NN

google-research/meta-dataset cambridge-mlg/cnaps

Learning to Compare: Relation Network for Few-Shot Learning

2017

Relation Networks

sicara/easy-few-shot-learning floodsung/LearningToCompare_FSL

Meta-Dataset

Performance Over Time

Edit Benchmark Results

Edit Result

Top Performing Models

All Papers (22)

Unleashing the Power of Meta-tuning for Few-shot Generalization Through Sparse Interpolated Experts

Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and Fine-Tuning Make a Difference

Task-Specific Preconditioner for Cross-Domain Few-Shot Learning

Cross-domain Few-shot Learning with Task-specific Adapters

Contextual Squeeze-and-Excitation for Efficient Few-Shot Image Classification

Universal Representation Learning from Multiple Domains for Few-shot Classification

Contextual Squeeze-and-Excitation for Efficient Few-Shot Image Classification

Shallow Bayesian Meta Learning for Real-World Few-Shot Recognition

A Universal Representation Transformer Layer for Few-Shot Image Classification

Selecting Relevant Features from a Multi-domain Representation for Few-shot Classification

Enhancing Few-Shot Image Classification with Unlabelled Examples

Improved Few-Shot Visual Classification

Selecting Relevant Features from a Multi-domain Representation for Few-shot Classification

Exploring Complementary Strengths of Invariant and Equivariant Representations for Few-Shot Learning

Fast and Flexible Multi-Task Classification Using Conditional Neural Adaptive Processes

Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples

Prototypical Networks for Few-shot Learning

Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples

Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks

Matching Networks for One Shot Learning

Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples

Learning to Compare: Relation Network for Few-Shot Learning

Model	Paper	Accuracy	Date
SMAT (DINO-VIT-Base-16-224)	Unleashing the Power of Meta-tuning for Few-shot …	85.27	2024-03-13
P>M>F (P=DINO-ViT-base, M=ProtoNet)	Pushing the Limits of Simple Pipelines for Few-Sh…	84.75	2022-04-15
TSP (ResNet18; applied on TA^2-Net)	Task-Specific Preconditioner for Cross-Domain Few…	81.40	2024-12-20
TSA (ResNet18, URL, residual adapters, 84x84 image, shuffled data, scratch, MDL)	Cross-domain Few-shot Learning with Task-specific…	78.07	2021-07-01
UpperCaSE-EfficientNetB0	Contextual Squeeze-and-Excitation for Efficient F…	76.10	2022-06-20
URL (ResNet18, 84x84 image, shuffled data, scratch, MDL)	Universal Representation Learning from Multiple D…	75.75	2021-03-25
UpperCaSE-ResNet50	Contextual Squeeze-and-Excitation for Efficient F…	74.90	2022-06-20
URT+MQDA	Shallow Bayesian Meta Learning for Real-World Few…	74.30	2021-01-08
URT	A Universal Representation Transformer Layer for …	72.15	2020-06-21
SUR	Selecting Relevant Features from a Multi-domain R…	70.72	2020-03-20
Transductive CNAPS	Enhancing Few-Shot Image Classification with Unla…	70.32	2020-06-17
Simple CNAPS	Improved Few-Shot Visual Classification	69.86	2019-12-07
SUR-pnf	Selecting Relevant Features from a Multi-domain R…	69.30	2020-03-20
Invariance-Equivariance	Exploring Complementary Strengths of Invariant an…	68.89	2021-03-01
CNAPs	Fast and Flexible Multi-Task Classification Using…	66.90	2019-06-18
fo-Proto-MAML	Meta-Dataset: A Dataset of Datasets for Learning …	63.43	2019-03-07
Prototypical Networks	Prototypical Networks for Few-shot Learning	60.57	2017-03-15
Finetune	Meta-Dataset: A Dataset of Datasets for Learning …	58.76	2019-03-07
fo-MAML	Model-Agnostic Meta-Learning for Fast Adaptation …	57.02	2017-03-09
Matching Networks	Matching Networks for One Shot Learning	56.25	2016-06-13
k-NN	Meta-Dataset: A Dataset of Datasets for Learning …	54.32	2019-03-07
Relation Networks	Learning to Compare: Relation Network for Few-Sho…	53.32	2017-11-16