YOLOV3+window10+cuda8.0+cudnn6.0安装过程

YOLOV3+window10+cuda8.0+cudnn6.0安装过程

MSVS 2015, CUDA 8.0 and OpenCV 3.0以上版本，

1、下载YOLOV3源码（github有人已经搭好的直接用） 
https://github.com/AlexeyAB/darknet 

把opencv拷贝到目录： C:\opencv_3.0\opencv\build\include & C:\opencv_3.0\opencv\build\x64\vc14\lib), then start MSVS, open build\darknet\darknet.sln, set x64 and Release, and do the: Build -> Build darknet. NOTE: If installing OpenCV, use OpenCV 3.4.0 or earlier. This is a bug in OpenCV 3.4.1 in the C API (see #500).

1.1. Find files opencv_world320.dll and opencv_ffmpeg320_64.dll (or opencv_world320.dll) in C:\opencv_3.0\opencv\build\x64\vc14\bin and put it near with darknet.exe

打开 build\darknet\darknet.vcxproj by using Notepad, find 2 places with "CUDA 9.1" 换成CUDA 8.0

然后编译darknet，
2、cuda8.0+cudann6.0安装参考博客 
http://blog.csdn.net/ezhchai/article/details/78411970 
3、配置opencv(本人配置的vs2105+opencv3.2.0)。 
4、进行编辑（debug及realse）。 
5、在目录底下“darknet-master\build\darknet\x64”会生成一个darknet.exe文件 
6、将opencv中的.dll文件拷到该目录底下:

opencv_world320.dll

7、下载权重yolov3.weights,下载后放在exe同级目录下，
https://pjreddie.com/darknet/yolo/ 
8、测试

darknet detect cfg/yolov3.cfg yolov3.weights data/dog.jpg

gtx 1070下， 测试320*320，需要40ms左右。