Awesome

Sphereface-model

This is the pre-trained model of SphereFace : Deep Hypersphere Embedding for Face Recognition. This model is trained on CASIA-Webface and the accuracy on LFW is 99.18%.

Training on Webface with the default setting only got an accuracy of 98.5%， which is much lower than the paper claimed(~99.26%).
Fixed lambda = 5, kept on training, and got an accuracy of 98.8%.
Decreased the lambda. Final accuracy is 99.18%.

I think the performance can still be further improved by carefully fine-tuning. Feel free to use this model.

Experiment Results

Train 22000(lambda=3.6, batch_size=170).

The distribution of features on LFW:

Feature distribution

The roc curve:

ROC

Accuracy on LFW:

Original	With PCA	With mirror trick	With mirror trick and PCA
98.88%	99.13%	98.98%	99.18%

Here is the model: https://pan.baidu.com/s/1pL0pmll

LFW evaluation code can be found in [https://github.com/happynear/FaceVerification].

Update

Accuracy on LFW 99.18% 2017.07.31

Reference

https://github.com/wy1iu/sphereface

https://github.com/happynear/NormFace/blob/master/MirrorFace.md