【官方框架地址】
https://github.com/ultralytics/ultralytics
【openvino介绍】
OpenVINO是一个针对Intel硬件优化的开源工具包,用于优化和部署深度学习模型。以下是OpenVINO部署模型的主要优点:
- 高性能:OpenVINO提供了一系列性能优化工具,如模型量化和剪枝等,可以在Intel硬件平台上实现高性能和低延迟的推理。
- 多平台支持:OpenVINO支持多种Intel硬件平台,包括CPU、GPU和FPGA等,可以在不同的硬件平台上进行优化和部署。
- 多框架支持:OpenVINO支持多种流行的深度学习框架,如TensorFlow、PyTorch和Caffe等,可以方便地将已有的模型集成到OpenVINO中。
- 简化部署:OpenVINO提供了一系列的工具和API,可以简化模型的部署和集成,使得开发者可以更快速地将模型部署到生产环境中。
- 预训练模型:OpenVINO提供了一系列的预训练模型,可以直接使用这些模型进行推理,也可以在这些模型的基础上进行微调和优化。
- 灵活性:OpenVINO提供了灵活的部署选项,可以将模型部署到本地设备、云端或边缘设备上,以满足不同的应用场景需求。
- 丰富的文档和社区支持:OpenVINO作为一个开源项目,拥有丰富的文档和活跃的社区支持,方便开发者学习和解决问题。
- 易用性:OpenVINO的接口简洁明了,易于使用,可以快速上手进行模型部署。
综上所述,OpenVINO在部署模型方面具有高性能、多平台支持、多框架支持、简化部署、预训练模型、灵活性、丰富的文档和社区支持以及易用性等优点。这些优点使得OpenVINO成为了一个强大而可靠的深度学习模型部署工具。
【效果展示】
【实现部分代码】
using System;
using System.Collections.Generic;
using System.ComponentModel;
using System.Data;
using System.Diagnostics;
using System.Drawing;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
using System.Windows.Forms;
using OpenCvSharp;
namespace FIRC
{
public partial class Form1 : Form
{
Mat src = new Mat();
Yolov8ClsManager detector = new Yolov8ClsManager();
public Form1()
{
InitializeComponent();
}
private void button1_Click(object sender, EventArgs e)
{
OpenFileDialog openFileDialog = new OpenFileDialog();
openFileDialog.Filter = "图文件(*.*)|*.jpg;*.png;*.jpeg;*.bmp";
openFileDialog.RestoreDirectory = true;
openFileDialog.Multiselect = false;
if (openFileDialog.ShowDialog() == DialogResult.OK)
{
src = Cv2.ImRead(openFileDialog.FileName);
pictureBox1.Image = OpenCvSharp.Extensions.BitmapConverter.ToBitmap(src);
}
}
private void button2_Click(object sender, EventArgs e)
{
if(pictureBox1.Image==null)
{
return;
}
var result = detector.Inference(src);
var resultMat = detector.DrawImage(result,src);
pictureBox2.Image= OpenCvSharp.Extensions.BitmapConverter.ToBitmap(resultMat); //Mat转Bitmap
}
private void Form1_Load(object sender, EventArgs e)
{
detector.LoadWeights(Application.StartupPath+ "\\weights\\yolov8n-cls.xml");
}
private void button3_Click(object sender, EventArgs e)
{
VideoCapture capture = new VideoCapture(0);
if (!capture.IsOpened())
{
Console.WriteLine("video not open!");
return;
}
Mat frame = new Mat();
var sw = new Stopwatch();
int fps = 0;
while (true)
{
capture.Read(frame);
if (frame.Empty())
{
Console.WriteLine("data is empty!");
break;
}
sw.Start();
var result = detector.Inference(src);
var resultMat = detector.DrawImage(result, src);
sw.Stop();
fps = Convert.ToInt32(1 / sw.Elapsed.TotalSeconds);
sw.Reset();
Cv2.PutText(resultMat, "FPS=" + fps, new OpenCvSharp.Point(30, 30), HersheyFonts.HersheyComplex, 1.0, new Scalar(255, 0, 0), 3);
//显示结果
Cv2.ImShow("Result", resultMat);
int key = Cv2.WaitKey(10);
if (key == 27)
break;
}
capture.Release();
}
}
}
【视频演示】
https://www.bilibili.com/video/BV1RK4y1q7qX/?vd_source=989ae2b903ea1b5acebbe2c4c4a635ee
【测试环境】
vs2019,netframework4.7.2