openvino最简单的部署

1.前言

• openvino可以看成一个驱动，你只要找到合适的头文件和库，就可以使用相应的设备，如CPU、GPU、计算棒等等。
• 当然，你必须先有一个安装了openvino的设备，参考这里
• 直通车，下载越早，花费越少

2.头文件和库

• 头文件
  • /opt/intel/openvino/deployment_tools/inference_engine/include
  • /opt/intel/openvino/deployment_tools/ngraph/include
  • /opt/intel/openvino/deployment_tools/inference_engine/samples/cpp/common/samples
• 库
  • /opt/intel/openvino/deployment_tools/inference_engine/lib/intel64
  • /opt/intel/openvino/deployment_tools/inference_engine/external/tbb/lib
  • /opt/intel/openvino/deployment_tools/ngraph/lib
  • 后面两个我只复制了libtbb.so.2和libngraph.so

3.C++代码

#include<string> 
#include<inference_engine.hpp> 
#include<samples/ocv_common.hpp>//matU8ToBlob
#include <ngraph/ngraph.hpp>
#include<time.h>
#include <sys/time.h>
#include <iostream>

using namespace InferenceEngine;

//将OpenCV Mat对象中的图像数据传给为InferenceEngine Blob对象
void frameToBlob(const cv::Mat& frame, InferRequest::Ptr& inferRequest, const std::string& inputName) {
   
   
	/* 从OpenCV Mat对象中拷贝图像数据到InferenceEngine 输入Blob对象 */
	Blob::Ptr frameBlob = inferRequest->GetBlob(inputName);
	matU8ToBlob<uint8_t>(frame, frameBlob);
}
//配置推理计算设备，IR文件路径，图片路径，阈值
std::string DEVICE = "MYRIAD";
std::string IR_FileNameNoExt = "/home/lwd/openvino/models/32/mobilenet-ssd";
std::string imageFile = "/home/lwd/openvino/mmgg.jpg";
float confidence_threshold = 0.07; 

int main(void) {
   
   
	// --------------------------- 1. 载入硬件插件(Plugin) --------------------------------------
	Core ie;
	std::cout << "1.Load Plugin..." << std::endl;     
	std::cout << ie.GetVersions(DEVICE) << std::endl; //输出插件版本信息

	// ------------------- 2. 读取IR文件 (.xml and .bin files) --------------------------
	std::cout << "2.Read IR File..." << std::endl;
	CNNNetwork network = ie.ReadNetwork(IR_FileNameNoExt+".xml");

	// -------------------- 3. 配置网络输入输出 -----------------------------------------
	std::cout << "3.Prepare Input and Output..." << std::endl;
	/** 获得神经网络的输入信息 **/
	InputsDataMap inputsInfo(network.getInputsInfo()