Omniglot

Introduced by Lake et al. in The Omniglot challenge: a 3-year progress report

The Omniglot data set is designed for developing more human-like learning algorithms. It contains 1623 different handwritten characters from 50 different alphabets. Each of the 1623 characters was drawn online via Amazon's Mechanical Turk by 20 different people. Each image is paired with stroke data, a sequences of [x,y,t] coordinates with time (t) in milliseconds.

Homepage

Benchmarks

Add a new result Link an existing benchmark

Task	Dataset Variant	Best Model
Few-Shot Image Classification	OMNIGLOT - 1-Shot, 20-way	GCR
Few-Shot Image Classification	OMNIGLOT - 5-Shot, 20-way	MC2+
Few-Shot Image Classification	OMNIGLOT - 1-Shot, 5-way	MC2+
Few-Shot Image Classification	OMNIGLOT - 5-Shot, 5-way	DCN6-E
Density Estimation	OMNIGLOT	B-NAF
One-Shot Segmentation	Cluttered Omniglot	MaskNet
Multi-Task Learning	OMNIGLOT	Gumbel-Matrix Routing
Few-Shot Image Classification	OMNIGLOT-EMNIST 5-way (1-shot)	HyperShot
Few-Shot Image Classification	OMNIGLOT-EMNIST 5-way (5-shot)	HyperShot
Personalized Federated Learning	Omniglot	pFedHN-PC
Few-Shot Image Classification	OMNIGLOT - 5-Shot, 423 way	APL
Few-Shot Image Classification	OMNIGLOT - 1-Shot, 1000 way	APL
Few-Shot Image Classification	OMNIGLOT - 5-Shot, 1000 way	APL
Meta-Learning	OMNIGLOT - 1-Shot, 20-way	Metadrop
Few-Shot Image Classification	OMNIGLOT - 1-Shot, 423 way	APL