NormFace: L2 Hypersphere Embedding for Face Verification

该文采用将特征或者权重矩阵归一化的方法，并引入了agent vecotor 代替了hard mining。论文中将方法在别人模型上进行finetuning，比自己从头训模型的效果要更好。

用来finetuning的模型：

Wu’s model is a 10-layer plain CNN with Maxout[6] activation unit.

https://github.com/AlfredXiangWu/face_verification_experiment

Wen’s model is a 28-layer ResNet[7] trained with both softmax loss and center loss.

https://github.com/ydwen/caffe-face 

论文代码：https://github.com/happynear/NormFace