Movatterモバイル変換


[0]ホーム

URL:


Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Sign up
Appearance settings

加入关键点的darknet训练框架,轻量级的人脸检测,支持ncnn推理

License

NotificationsYou must be signed in to change notification settings

DayBreak-u/darknet_face_with_landmark

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

32 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

借鉴AlexeyAB大神的darknet 做适量修改,用于人脸检测以及关键点检测,支持ncnn推理

实现的功能

  • 添加关键点检测分支,使用wing loss
  • 添加 hswish,hsigmode 激活函数

Installation

Clone and install
  1. git clonehttps://github.com/ouyanghuiyu/darknet_face_with_landmark.git
  2. 使用scripts/retinaface2yololandmark.py脚本将retinaface的标记文件转为yolo的格式使用
  3. 其他编译训练都和原版darknet相同
  4. 测试
    ./darknet detector test ./data/face.data  ./cfg/mbv2_yolov3_face.cfg  ./models/mbv2_yolov3_face_final.weights  ./test_imgs/input/selfie.jpg  -dont_show

或者使用yolo_landmark.py进行测试,更换里面的模型配置文件即可

精度

Widerface测试

  • 在wider face val精度(单尺度输入分辨率:320*240
方法EasyMediumHard
libfacedetection v1(caffe)0.650.50.233
libfacedetection v2(caffe)0.7140.5850.306
Retinaface-Mobilenet-0.25(Mxnet)0.7450.5530.232
mbv2_yolov3_face(our)0.840.790.41
  • 在wider face val精度(单尺度输入分辨率:640*480
方法EasyMediumHard
libfacedetection v1(caffe)0.7410.6830.421
libfacedetection v2(caffe)0.7730.7180.485
Retinaface-Mobilenet-0.25(Mxnet)0.8790.8070.481
mbv2_yolov3_face(our)0.8660.8480.718

ps: 测试的时候,长边为320 或者 640 ,图像等比例缩放,yolo未作缩放.

测试

References

About

加入关键点的darknet训练框架,轻量级的人脸检测,支持ncnn推理

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

[8]ページ先頭

©2009-2025 Movatter.jp