Darknet Yolov4安装小记

基本配置

opencv 4.5.1
cuda 11.2
cudnn 8.1

下载darknet(Yolov4)

git clone https://github.com/AlexeyAB/darknet.git

安装

cd darknet/
make
./build.sh (这步很多教程没有，但尤为重要，由它生成py文件需要的libdarknet.so)

测试

//下载yolov4.weight放在文件根目录
https://github.com/AlexeyAB/darknet/releases/download/darknet_yolo_v3_optimal/yolov4.weights
//yolov4-tiny.weights（选用）
https://github.com/AlexeyAB/darknet/releases/download/darknet_yolo_v4_pre/yolov4-tiny.weights

//image
./darknet detect cfg/yolov4.cfg yolov4.weights data/dog.jpg 
//如果执行后长时间没结果，我这边的解决方法是将CMakeLists.txt
//中 option(ENABLE_CUDNN_HALF "Enable CUDNN Half precision" ON) 改为OFF，
//执行 make clean
//删除build_release目录
//重新执行安装

//video
./darknet detector demo cfg/coco.data cfg/yolov4.cfg yolov4.weights test.mp4 -out_filename res.avi

//webcam
./darknet detector demo cfg/coco.data cfg/yolov4.cfg yolov4.weights -c 0

//测试python
python3 darknet_images.py