/MobileFaceNet_TF

Tensorflow implementation for MobileFaceNet

Primary LanguagePythonApache License 2.0Apache-2.0

MobileFaceNet_TF

Tensorflow implementation for MobileFaceNet.

dependencies

  • tensorflow >= r1.5
  • opencv-python 3.x
  • python 3.x
  • scipy
  • sklearn
  • numpy
  • mxnet
  • pickle

Prepare dataset

  1. choose one of the following links to download dataset which is provide by insightface. (Special Recommend MS1M-refine-v2)
  1. move dataset to ${MobileFaceNet_TF_ROOT}/datasets.
  2. run ${MobileFaceNet_TF_ROOT}/utils/data_process.py.

pretrained model

training

  1. refined super parameters by yourself special project.
  2. run script ${MobileFaceNet_TF_ROOT}/train_nets.py
  3. have a snapshot result at ${MobileFaceNet_TF_ROOT}/output.

performance

size LFW(%) Val@1e-3(%) inference@MSM8976-cpu(ms)
5.7M 99.4+ 98.4+ 260-

References

  1. facenet
  2. InsightFace mxnet
  3. InsightFace_TF
  4. MobileFaceNets: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devices
  5. CosFace: Large Margin Cosine Loss for Deep Face Recognition
  6. InsightFace : Additive Angular Margin Loss for Deep Face Recognition