YOLO-FaceV2

Introduction

YOLO-FaceV2: A Scale and Occlusion Aware Face Detector
https://arxiv.org/abs/2208.02019

Framework Structure

Environment Requirments

Create a Python Virtual Environment.

conda create -n {name} python=x.x

Enter Python Virtual Environment.

conda activate {name}

Install pytorch in this.

pip install torch==1.10.0+cu111 torchvision==0.11.0+cu111 torchaudio==0.10.0 -f https://download.pytorch.org/whl/torch_stable.html

Install other python package.

pip install -r requirements.txt

Step-Through Example

Installation

Get the code.

git clone https://github.com/Krasjet-Yu/YOLO-FaceV2.git

Dataset

Download the WIDER FACE dataset. Then convert it to YOLO format.

# You can modify convert.py and voc_label.py if needed.
python3 data/convert.py
python3 data/voc_label.py

Preweight

link: https://pan.baidu.com/s/1FVIY20qtTSM9gDhz7DtJkA

code: tzfs

Training

Train your model on WIDER FACE.

python train.py --weights preweight.pt    
                --data data/WIDER_FACE.yaml    
                --cfg models/yolov5s_v2_RFEM_MultiSEAM.yaml     
                --batch-size 64   
                --epochs 250

Evaluate

Evaluate the trained model via next code on WIDER FACE

python widerface_pred.py --weights runs/train/x/weights/best.pt     
                         --save_folder ./widerface_evaluate/widerface_txt_x    
cd widerface_evaluate/    
python evaluation.py --pred ./widerface_txt_x

Download the eval_tool to show the performance.

The result is shown below:

Finetune

see in ultralytics/yolov5#607

# Single-GPU
python train.py --epochs 10 --data coco128.yaml --weights yolov5s.pt --cache --evolve

# Multi-GPU
for i in 0 1 2 3 4 5 6 7; do
  sleep $(expr 30 \* $i) &&  # 30-second delay (optional)
  echo 'Starting GPU '$i'...' &&
  nohup python train.py --epochs 10 --data coco128.yaml --weights yolov5s.pt --cache --device $i --evolve > evolve_gpu_$i.log &
done

# Multi-GPU bash-while (not recommended)
for i in 0 1 2 3 4 5 6 7; do
  sleep $(expr 30 \* $i) &&  # 30-second delay (optional)
  echo 'Starting GPU '$i'...' &&
  "$(while true; do nohup python train.py... --device $i --evolve 1 > evolve_gpu_$i.log; done)" &
done

Reference

https://github.com/ultralytics/yolov5

https://github.com/deepcam-cn/yolov5-face

https://github.com/open-mmlab/mmdetection

https://github.com/dongdonghy/repulsion_loss_pytorch

cite

If you think this work is helpful for you, please cite

@ARTICLE{2022arXiv220802019Y,
       author = {{Yu}, Ziping and {Huang}, Hongbo and {Chen}, Weijun and {Su}, Yongxin and {Liu}, Yahui and {Wang}, Xiuying},
        title = "{YOLO-FaceV2: A Scale and Occlusion Aware Face Detector}",
      journal = {arXiv e-prints},
     keywords = {Computer Science - Computer Vision and Pattern Recognition},
         year = 2022,
        month = aug,
          eid = {arXiv:2208.02019},
        pages = {arXiv:2208.02019},
archivePrefix = {arXiv},
       eprint = {2208.02019},
 primaryClass = {cs.CV},
       adsurl = {https://ui.adsabs.harvard.edu/abs/2022arXiv220802019Y},
      adsnote = {Provided by the SAO/NASA Astrophysics Data System}
}

Contact

We use code's license is MIT License. The code can be used for business inquiries or professional support requests.

WangDeyu/YOLO-FaceV2