BoN uses an online-updated binary hash table to sample high-quality negatives for Siamese network training, improving speed and accuracy for triplet loss.
Deep metric learning with hierarchical triplet loss
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Bag of Negatives for Siamese Architectures
BoN uses an online-updated binary hash table to sample high-quality negatives for Siamese network training, improving speed and accuracy for triplet loss.