//解压l_openvino_toolkit_p_2018.5.445.tgz
cd l_openvino_toolkit_p_2018.5.445
sudo -E ./install_cv_sdk_dependencies.sh
sudo ./install_GUI.sh
//出现图形界面,然后一路next就好
//添加环境变量
sudo gedit ~/.bashrc
source /opt/intel/computer_vision_sdk/bin/setupvars.sh
source ~/.bashrc
cd /opt/intel/computer_vision_sdk/deployment_tools/model_optimizer/install_prerequisites
sudo ./install_prerequisites.sh
//运行图像分类demo
cd /opt/intel/computer_vision_sdk/deployment_tools/demo
./demo_squeezenet_download_convert_run.sh
//运行inference pipeline demo
./demo_security_barrier_camera.sh
1,将keras模型转tensorflow模型,.h5转.pb:
链接:https://github.com/amir-abdi/keras_to_tensorflow
2,.pb转成.xml和.bin
具体操作可参考官方文档:https://software.intel.com/en-us/articles/OpenVINO-Using-TensorFlow
3,
安装过程出现的bug:
1,错误名称:Ubuntu:apt-get update出错:由于没有公钥,无法验证下列签名
解决办法:
sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys E40EBBA24FF2FC69
//注意修改公钥名称E40EBBA24FF2FC69
2,错误名称:[ ERROR ] Shape [ -1 224 224 3] is not fully defined for output 0 of “input_1”. Use --input_shape with positive integers to override model input shapes.
解决方法:
//加上
--input_shape [1,224,224,3]
//或者
-b 1