Abstract
Convolutional neural networks have significantly boosted the performance of
face recognition in recent years due to its high capacity in learning
discriminative features. To enhance the discriminative power of the Softmax
loss, multiplicative angular margin and additive cosine margin incorporate
angular margin and cosine margin into the loss functions, respectively. In this
paper, we propose a novel supervisor signal, additive angular margin (ArcFace),
which has a better geometrical interpretation than supervision signals proposed
so far. Specifically, the proposed ArcFace \$\cos(+ m)\$ directly maximise
decision boundary in angular (arc) space based on the L2 normalised weights and
features. Compared to multiplicative angular margin \$\cos(mþeta)\$ and
additive cosine margin \$\cosþeta-m\$, ArcFace can obtain more discriminative
deep features. We also emphasise the importance of network settings and data
refinement in the problem of deep face recognition. Extensive experiments on
several relevant face recognition benchmarks, LFW, CFP and AgeDB, prove the
effectiveness of the proposed ArcFace. Most importantly, we get state-of-art
performance in the MegaFace Challenge in a totally reproducible way. We make
data, models and training/test code public
available\~<a href="https://github.com/deepinsight/insightface">this https URL</a>.
Users
Please
log in to take part in the discussion (add own reviews or comments).