Darknet YOLOの進化がとまらない

github.com

YOLOで『densenet201』が使えるようになった。

  • pre-trained weights 『densenet201.300』をダウンロード
  • cfg-file 『densenet201_yolo2.cfg』をダウンロード
darknet.exe detector train obj.data densenet201_yolo2.cfg densenet201.300

(おまけ)Yolo_markの使い方

github.com
To compile on Windows, open in MSVS2015 yolo_mark.sln and compile it.
1) If you have other version of OpenCV 2.4.9 then you should change version number in first 3 lines in main.cpp.
2) If you have OpenCV 2.4.13 then you should change pathes

  • 2-1) (right click on project) -> properties -> C/C++ -> General -> Additional Include Directories: C:\opencv_2.4.13\opencv\build\include

  • 2-2) (right click on project) -> properties -> Linker -> General -> Additional Library Directories: C:\opencv_2.4.13\opencv\build\x64\vc14\lib