Multiplying a face-recognition loss by a sigmoid of that loss, which up-weights hard examples and down-weights easy ones, gives small and inconsistent accuracy changes on LFW and YTF.
IEEE Signal Processing Letters 25(5), 635–639 (2018)
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
-
Hard-Mining Loss based Convolutional Neural Network for Face Recognition
Multiplying a face-recognition loss by a sigmoid of that loss, which up-weights hard examples and down-weights easy ones, gives small and inconsistent accuracy changes on LFW and YTF.