Few-Shot Image Classification on Meta-Dataset

Metric: Accuracy (higher is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	Accuracy▼	Extra Data	Paper	Date↕	Code
1	SMAT (DINO-VIT-Base-16-224)	85.27	No	Unleashing the Power of Meta-tuning for Few-shot...	2024-03-13	Code
2	P>M>F (P=DINO-ViT-base, M=ProtoNet)	84.75	No	Pushing the Limits of Simple Pipelines for Few-S...	2022-04-15	Code
3	TSP (ResNet18; applied on TA^2-Net)	81.4	No	Task-Specific Preconditioner for Cross-Domain Fe...	2024-12-20	-
4	TSA (ResNet18, URL, residual adapters, 84x84 image, shuffled data, scratch, MDL)	78.07	No	Cross-domain Few-shot Learning with Task-specifi...	2021-07-01	Code
5	UpperCaSE-EfficientNetB0	76.1	No	Contextual Squeeze-and-Excitation for Efficient ...	2022-06-20	Code
6	URL (ResNet18, 84x84 image, shuffled data, scratch, MDL)	75.75	No	Universal Representation Learning from Multiple ...	2021-03-25	Code
7	UpperCaSE-ResNet50	74.9	No	Contextual Squeeze-and-Excitation for Efficient ...	2022-06-20	Code
8	URT+MQDA	74.3	No	Shallow Bayesian Meta Learning for Real-World Fe...	2021-01-08	Code
9	URT	72.15	No	A Universal Representation Transformer Layer for...	2020-06-21	Code
10	SUR	70.72	No	Selecting Relevant Features from a Multi-domain ...	2020-03-20	Code
11	Transductive CNAPS	70.32	No	Enhancing Few-Shot Image Classification with Unl...	2020-06-17	Code
12	Simple CNAPS	69.86	No	Improved Few-Shot Visual Classification	2019-12-07	Code
13	SUR-pnf	69.3	No	Selecting Relevant Features from a Multi-domain ...	2020-03-20	Code
14	Invariance-Equivariance	68.89	No	Exploring Complementary Strengths of Invariant a...	2021-03-01	Code
15	CNAPs	66.9	No	Fast and Flexible Multi-Task Classification Usin...	2019-06-18	Code
16	fo-Proto-MAML	63.428	No	Meta-Dataset: A Dataset of Datasets for Learning...	2019-03-07	Code
17	Prototypical Networks	60.573	No	Prototypical Networks for Few-shot Learning	2017-03-15	Code
18	Finetune	58.758	No	Meta-Dataset: A Dataset of Datasets for Learning...	2019-03-07	Code
19	fo-MAML	57.024	No	Model-Agnostic Meta-Learning for Fast Adaptation...	2017-03-09	Code
20	Matching Networks	56.247	No	Matching Networks for One Shot Learning	2016-06-13	Code
21	k-NN	54.319	No	Meta-Dataset: A Dataset of Datasets for Learning...	2019-03-07	Code
22	Relation Networks	53.315	No	Learning to Compare: Relation Network for Few-Sh...	2017-11-16	Code

#1SMAT (DINO-VIT-Base-16-224)SOTA
85.27
Accuracy· 2024-03-13
Unleashing the Power of Meta-tuning for Few-shot Generalization Through Sparse Interpolated Experts Code
#2P>M>F (P=DINO-ViT-base, M=ProtoNet)SOTA
84.75
Accuracy· 2022-04-15
Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and Fine-Tuning Make a Difference Code
#3TSP (ResNet18; applied on TA^2-Net)
81.4
Accuracy· 2024-12-20
Task-Specific Preconditioner for Cross-Domain Few-Shot Learning
#4TSA (ResNet18, URL, residual adapters, 84x84 image, shuffled data, scratch, MDL)SOTA
78.07
Accuracy· 2021-07-01
Cross-domain Few-shot Learning with Task-specific Adapters Code
#5UpperCaSE-EfficientNetB0
76.1
Accuracy· 2022-06-20
Contextual Squeeze-and-Excitation for Efficient Few-Shot Image Classification Code
#6URL (ResNet18, 84x84 image, shuffled data, scratch, MDL)SOTA
75.75
Accuracy· 2021-03-25
Universal Representation Learning from Multiple Domains for Few-shot Classification Code
#7UpperCaSE-ResNet50
74.9
Accuracy· 2022-06-20
Contextual Squeeze-and-Excitation for Efficient Few-Shot Image Classification Code
#8URT+MQDASOTA
74.3
Accuracy· 2021-01-08
Shallow Bayesian Meta Learning for Real-World Few-Shot Recognition Code
#9URTSOTA
72.15
Accuracy· 2020-06-21
A Universal Representation Transformer Layer for Few-Shot Image Classification Code
#10SURSOTA
70.72
Accuracy· 2020-03-20
Selecting Relevant Features from a Multi-domain Representation for Few-shot Classification Code
#11Transductive CNAPS
70.32
Accuracy· 2020-06-17
Enhancing Few-Shot Image Classification with Unlabelled Examples Code
#12Simple CNAPSSOTA
69.86
Accuracy· 2019-12-07
Improved Few-Shot Visual Classification Code
#13SUR-pnf
69.3
Accuracy· 2020-03-20
Selecting Relevant Features from a Multi-domain Representation for Few-shot Classification Code
#14Invariance-Equivariance
68.89
Accuracy· 2021-03-01
Exploring Complementary Strengths of Invariant and Equivariant Representations for Few-Shot Learning Code
#15CNAPsSOTA
66.9
Accuracy· 2019-06-18
Fast and Flexible Multi-Task Classification Using Conditional Neural Adaptive Processes Code
#16fo-Proto-MAMLSOTA
63.428
Accuracy· 2019-03-07
Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples Code
#17Prototypical NetworksSOTA
60.573
Accuracy· 2017-03-15
Prototypical Networks for Few-shot Learning Code
#18Finetune
58.758
Accuracy· 2019-03-07
Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples Code
#19fo-MAMLSOTA
57.024
Accuracy· 2017-03-09
Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks Code
#20Matching NetworksSOTA
56.247
Accuracy· 2016-06-13
Matching Networks for One Shot Learning Code
#21k-NN
54.319
Accuracy· 2019-03-07
Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples Code
#22Relation Networks
53.315
Accuracy· 2017-11-16
Learning to Compare: Relation Network for Few-Shot Learning Code