- Notifications
You must be signed in to change notification settings - Fork70
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
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
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
- 添加关键点检测分支,使用wing loss
- 添加 hswish,hsigmode 激活函数
- git clonehttps://github.com/ouyanghuiyu/darknet_face_with_landmark.git
- 使用scripts/retinaface2yololandmark.py脚本将retinaface的标记文件转为yolo的格式使用
- 其他编译训练都和原版darknet相同
- 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
- 在wider face val精度(单尺度输入分辨率:320*240)
方法 | Easy | Medium | Hard |
---|---|---|---|
libfacedetection v1(caffe) | 0.65 | 0.5 | 0.233 |
libfacedetection v2(caffe) | 0.714 | 0.585 | 0.306 |
Retinaface-Mobilenet-0.25(Mxnet) | 0.745 | 0.553 | 0.232 |
mbv2_yolov3_face(our) | 0.84 | 0.79 | 0.41 |
- 在wider face val精度(单尺度输入分辨率:640*480)
方法 | Easy | Medium | Hard |
---|---|---|---|
libfacedetection v1(caffe) | 0.741 | 0.683 | 0.421 |
libfacedetection v2(caffe) | 0.773 | 0.718 | 0.485 |
Retinaface-Mobilenet-0.25(Mxnet) | 0.879 | 0.807 | 0.481 |
mbv2_yolov3_face(our) | 0.866 | 0.848 | 0.718 |
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.
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
Uh oh!
There was an error while loading.Please reload this page.
Stars
Watchers
Forks
Releases
No releases published
Packages0
No packages published
Uh oh!
There was an error while loading.Please reload this page.