Repository files navigation

GluonFR is a toolkit based on MXnet-Gluon, provides SOTA deep learning algorithm and models in face recognition.

GluonFR supports Python 3.5 or later.To install this package you need install GluonCV and MXNet first:

pip install gluoncv --prepip install mxnet-mkl --pre --upgrade# if cuda XX is installedpip install mxnet-cuXXmkl --pre --upgrade

Then install gluonfr:

• From Source(recommend)

pip install git+https://github.com/THUFutureLab/gluon-face.git@master

• Pip

pip install gluonfr

GluonFR is based on MXnet-Gluon, if you are new to it, please check outdmlc 60-minute crash course.

This part provides input pipeline for training and validation,all datasets is aligned by mtcnn and cropped to (112, 112) by DeepInsight,they converted images totrain.rec,train.idx andval_data.bin files, please check out[insightface/Dataset-Zoo] for more information.Indata/dali_utils.py, there is a simple example of Nvidia-DALI. It is worth trying when data augmentation with cpucan not satisfy the speed of gpu training,

The files should be prepared like:

face/    emore/        train.rec        train.idx        property    ms1m/        train.rec        train.idx        property    lfw.bin    agedb_30.bin    ...    vgg2_fp.bin

We use~/.mxnet/datasets as default dataset root to match mxnet setting.

mobile_facenet, res_attention_net, se_resnet...

GluonFR provides implement of losses in recent, including SoftmaxCrossEntropyLoss, ArcLoss, TripletLoss,RingLoss, CosLoss, L2Softmax, ASoftmax, CenterLoss, ContrastiveLoss, ... , and we will keep updating in future.
If there is any method we overlooked, please open anissue.

examples/ shows how to use gluonfr to train a face recognition model, and how to get Mnist 2-Dfeature embedding visualization.

The last column of this chart is the best LFW accuracy reported in paper, they are trained with different data and networks,later we will give our results of these method with same train data and network.

Method	Paper	Visualization of MNIST	LFW
Contrastive Loss	ContrastiveLoss	-	-
Triplet	1503.03832	-	99.63±0.09
Center Loss	CenterLoss		99.28
L2-Softmax	1703.09507	-	99.33
A-Softmax	1704.08063	-	99.42
CosLoss/AMSoftmax	1801.05599/1801.05599		99.17
Arcloss	1801.07698		99.82
Ring loss	1803.00130		99.52
LGM Loss	1803.02988		99.20±0.03

SeeModel Zoo in doc.

• More pretrained models
• IJB and Megaface Results
• Other losses
• Dataloader for loss depend on how to provide batches like Triplet, ContrastiveLoss, RangeLoss...
• Try GluonCV resnetV1b/c/d/ to improve performance
• Create hosted docs
• Test module
• Pypi package

Please checkoutlink.
For Chinese Version:link

{haoxintongYangxvHaoyadongSunhao }

中文社区Gluon-Forum Feel free to use English here :D.

1. MXNet Documentation and Tutorialshttps://zh.diveintodeeplearning.org/

2. NVIDIA DALI documentationNVIDIA DALI documentation

3. Deepinsightinsightface