List your company
ResearchFormerly WizWand

Few-shot Character Recognition on Omniglot (novel classes)

99.53AccuracyMAML
94.704495.957297.2198.4628Apr 8, 2019
Updated 8mo ago

Evaluation results

MethodLinks
MAMLN-shot=5-shot, Backbone=Conv-4, Data Augmentation=None, N-way=5-way2019.0499.53
Baseline++N-shot=5-shot, Backbone=Conv-4, Data Augmentation=None, N-way=5-way, Training Epochs=52019.0499.38
MatchingNetN-shot=5-shot, Backbone=Conv-4, Data Augmentation=None, N-way=5-way2019.0499.37
RelationNetN-shot=5-shot, Backbone=Conv-4, Data Augmentation=None, N-way=5-way2019.0499.3
ProtoNetN-shot=5-shot, Backbone=Conv-4, Data Augmentation=None, N-way=5-way2019.0499.15
BaselineN-shot=5-shot, Backbone=Conv-4, Data Augmentation=None, N-way=5-way, Training Epochs=52019.0499.12
MAMLN-shot=1-shot, Backbone=Conv-4, Data Augmentation=None, N-way=5-way2019.0498.57
ProtoNetN-shot=1-shot, Backbone=Conv-4, Data Augmentation=None, N-way=5-way2019.0498.01
MatchingNetN-shot=1-shot, Backbone=Conv-4, Data Augmentation=None, N-way=5-way2019.0497.78
RelationNetN-shot=1-shot, Backbone=Conv-4, Data Augmentation=None, N-way=5-way2019.0497.22
Baseline++N-shot=1-shot, Backbone=Conv-4, Data Augmentation=None, N-way=5-way, Training Epochs=52019.0495.41
BaselineN-shot=1-shot, Backbone=Conv-4, Data Augmentation=None, N-way=5-way, Training Epochs=52019.0494.89