Skip to content

Facenet implemented by tensorflow2.0-keras api and in eager-mode support CenterLoss, LSoftmaxLoss, L2SoftmaxLoss, AMSoftmaxLoss, ASoftmaxLoss, ArcFaceSoftmaxLoss

Notifications You must be signed in to change notification settings

DequanZhu/FaceNet-and-FaceLoss-collections-tensorflow2.0

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

64 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Description

This repository implements facenet using tensorflow2.0-keras api and in eager-mode supporting different backbones and different loss types. The project is still undergo.

How to use

  • Install tensorflow2.0-beta1 version.
  • Anaconda virtual enviroment is recommended.
  • I use tfrecords format data to create input-pipeline.To create tfrecords format training data, run the script datasets.py by:
  bash scripts/tfrcd_data_gen.sh

TODO

  • Provide train code in vggface2 datasets using SoftmaxLoss.
  • To implement CenterLoss, LSoftmaxLoss, L2SoftmaxLoss, AMSoftmaxLoss, ASoftmaxLoss, ArcFaceSoftmaxLoss.
  • Provide test code in LFW datasets.
  • Train some models using different backbone and diffrent loss fun in secord.
  • Compare and analysis different loss type.
  • Refactor and clean the code.

About

Facenet implemented by tensorflow2.0-keras api and in eager-mode support CenterLoss, LSoftmaxLoss, L2SoftmaxLoss, AMSoftmaxLoss, ASoftmaxLoss, ArcFaceSoftmaxLoss

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published