InsightFace : Implementation for paper 'ArcFace: Additive Angular Margin Loss for Deep Face Recognition'

Paper by Jiankang Deng, Jia Guo, and Stefanos Zafeiriou

License

InsightFace is released under the MIT License.

Introduction
Citation
Requirements
Installation
Usage
Models
Results
Contribution
Contact

Introduction

The repository contains the entire pipeline for deep face recognition with ArcFace and other popular methods including Softmax, Triplet Loss, SphereFace and AMSoftmax/CosineFace, etc..

ArcFace is a recently proposed face recognition method. It was initially described in an arXiv technical report. By using ArcFace and this repository, you can simply achieve LFW 99.80+ and Megaface 98%+ by a single model.

We provide a refined MS1M dataset for training here, which was already packed in MXNet binary format. It allows researcher or industrial engineer to develop a deep face recognizer quickly by only two stages: 1. Download binary dataset; 2. Run training script.

In InsightFace, we support several popular network backbones and can be set just in one parameter. Below is the list until today:

ResNet
MobiletNet
InceptionResNetV2
DPN
DenseNet

In our paper, we found there're overlap identities between facescrub dataset and Megaface distractors which greatly affects the identification performance. Sometimes more than 10 percent improvement can be achieved after removing these overlaps. This list will be made public soon in this repository.

ArcFace achieves the state-of-the-art identification performance in MegaFace Challenge, at 98%+.

Citation

If you find InsightFace/ArcFace useful in your research, please consider to cite our paper.

Requirements

  1. The only requirement is `MXNet` with GPU support(Python 2.7).

Installation

Install MXNet by
```
pip install mxnet-cu80
```
Clone the InsightFace repository. We'll call the directory that you cloned InsightFace as INSIGHTFACE_ROOT.
```
git clone --recursive https://github.com/deepinsight/insightface.git
```

Usage

After successfully completing the installation, you are ready to run all the following experiments.

Part 1: Dataset Downloading.

Note: In this part, we assume you are in the directory $INSIGHTFACE_ROOT/

Download the training set (MS1M) from [here] and place them in datasets/. Each training dataset includes following 7 files:
```
	- train.idx
	- train.rec
	- property
	- lfw.bin
	- cfp_ff.bin
	- cfp_fp.bin
	- agedb_30.bin
```
The first three files are the dataset itself while the last four ones are binary verification sets.

Part 2: Train

Note: In this part, we assume you are in the directory $INSIGHTFACE_ROOT/src/. Before start any training procedure, make sure you set the correct env params for MXNet to ensure the performance.

export MXNET_CPU_WORKER_NTHREADS=24
export MXNET_ENGINE_TYPE=ThreadedEnginePerDevice

Now we give some examples below. Our experiments were all done on Tesla P40 GPU.

Train ArcFace with LResNet100E-IR.
```
CUDA_VISIBLE_DEVICES='0,1,2,3' python -u train_softmax.py --data-dir ../datasets/faces_ms1mr_112x112 --network r100 --loss-type 4 --prefix ../model-r100-arcface --margin-m 0.5
```
It will output verification results of LFW, CFP-FF, CFP-FP and AgeDB-30 every 2000 batches. You can check all command line options in train_softmax.py.

This model can achieve LFW 99.80+ and MegaFace 98.0%+

Train AMSoftmax/CosineFace with LResNet50E-IR.

CUDA_VISIBLE_DEVICES='0,1,2,3' python -u train_softmax.py --data-dir ../datasets/faces_ms1mr_112x112 --network r50 --loss-type 2 --prefix ../model-r50-amsoftmax --margin-m 0.35

Train Softmax with LMobileNetE.

CUDA_VISIBLE_DEVICES='0,1,2,3' python -u train_softmax.py --data-dir ../datasets/faces_ms1mr_112x112 --network m1 --loss-type 0 --prefix ../model-m1-softmax

Re-Train with Triplet on above Softmax model.

CUDA_VISIBLE_DEVICES='0,1,2,3' python -u train_softmax.py --data-dir ../datasets/faces_ms1mr_112x112 --network m1 --loss-type 12 --pretrained ../model-m1-softmax,50 --prefix ../model-m1-triplet --lr 0.005 --mom 0.0 --per-batch-size 150

Train Softmax with LDPN107E.
```
CUDA_VISIBLE_DEVICES='0,1,2,3,4,5,6,7' python -u train_softmax.py --data-dir ../datasets/faces_vgg_112x112 --network p107 --loss-type 0 --prefix ../model-p107-softmax --per-batch-size 64
```
Part 3: MegaFace Test

Note: In this part, we assume you are in the directory $INSIGHTFACE_ROOT/src/megaface/
1. Align all face images of facescrub dataset and megaface distractors. Please check the alignment scripts under $INSIGHTFACE_ROOT/src/align/. (We may plan to release these data soon, not sure.)
2. Next, generate feature files for both facescrub and megaface images.
```
python -u gen_megaface.py
```
3. Remove Megaface noises which generates new feature files.
```
python -u remove_noises.py
```
4. Start to run megaface development kit to produce final result.

Models

1. We plan to make some models public soon.

Results

We simply report the performance of LResNet100E-IR network trained on MS1M dataset with ArcFace loss.

Method	LFW(%)	CFP-FF(%)	CFP-FP(%)	AgeDB-30(%)	MegaFace1M(%)
ArcFace	99.80+	99.85+	94.0+	97.90+	98.0+

Contribution

Any type of PR or third-party contribution are welcome.

Contact

 [Jia Guo](guojia[at]gmail.com) and [Jiankang Deng](https://ibug.doc.ic.ac.uk/people/jdeng)

 Questions can also be left as issues in the repository. We will be happy to answer them.

Cv9527 / insightface