C# OpenCvSharp Yolov8 Pose gesture recognition

Table of Contents

Effect

project

Model information

code

download

---

Effect

project

VS2022

.net framework 4.8

OpenCvSharp 4.8

Microsoft.ML.OnnxRuntime 1.16.2

Model information

Model Properties
———————–
date:2023-09-07T17:11:43.091306
description: Ultralytics YOLOv8n-pose model trained on /usr/src/app/ultralytics/datasets/coco-pose.yaml
author:Ultralytics
kpt_shape：[17, 3]
task: pose
license: AGPL-3.0 https://ultralytics.com/license
version:8.0.172
stride: 32
batch: 1
imgsz：[640, 640]
names: {0: ‘person’}
————————————————– ————-

Inputs
———————–
name:images
tensor: Float[1, 3, 640, 640]
————————————————– ————-

Outputs
———————–
name:output0
tensor: Float[1, 56, 8400]
————————————————– ————-

code

//Zoom the picture
max_image_length = image.Cols > image.Rows ? image.Cols : image.Rows;
max_image = Mat.Zeros(new OpenCvSharp.Size(max_image_length, max_image_length), MatType.CV_8UC3);
roi = new Rect(0, 0, image.Cols, image.Rows);
image.CopyTo(new Mat(max_image, roi));

factors[0] = factors[1] = (float)(max_image_length / 640.0);

//Data normalization processing
BN_image = CvDnn.BlobFromImage(max_image, 1 / 255.0, new OpenCvSharp.Size(640, 640), new Scalar(0, 0, 0), true, false);

//Configure image input data
opencv_net.SetInput(BN_image);

dt1 = DateTime.Now;
//Model inference, read inference results
result_mat = opencv_net.Forward();
dt2 = DateTime.Now;

//Convert the inference result to float data type
result_mat_to_float = new Mat(8400, 56, MatType.CV_32F, result_mat.Data);

//Read data into array
result_mat_to_float.GetArray(out result_array);

result = result_pro.process_result(result_array);

result_image = result_pro.draw_result(result, image.Clone());

if (!result_image.Empty())
{
pictureBox2.Image = new Bitmap(result_image.ToMemoryStream());
sb.Clear();
sb.AppendLine(“Inference time consumption:” + (dt2 – dt1).TotalMilliseconds + “ms”);
sb.AppendLine(“———————————“);
textBox1.Text = sb.ToString();
}
else
{
textBox1.Text = “No information”;
}

using OpenCvSharp;
using OpenCvSharp.Dnn;
using System;
using System.Collections.Generic;
using System.ComponentModel;
using System.Data;
using System.Drawing;
using System.Linq;
using System.Text;
using System.Windows.Forms;

namespace OpenCvSharp_Yolov8_Demo
{
    public partial class Form1 : Form
    {
        public Form1()
        {
            InitializeComponent();
        }

        string fileFilter = "*.*|*.bmp;*.jpg;*.jpeg;*.tiff;*.tiff;*.png";
        string image_path = "";
        string startupPath;
        string classer_path;

        DateTime dt1 = DateTime.Now;
        DateTime dt2 = DateTime.Now;
        string model_path;
        Mat image;

        PoseResult result_pro;
        Mat result_mat;
        Mat result_image;
        Mat result_mat_to_float;

        Net opencv_net;
        Mat BN_image;

        float[] result_array;
        float[] factors;

        int max_image_length;
        Mat max_image;
        Rectroi;

        Result result;
        StringBuilder sb = new StringBuilder();

        private void Form1_Load(object sender, EventArgs e)
        {
            startupPath = System.Windows.Forms.Application.StartupPath;
            model_path = startupPath + "\yolov8n-pose.onnx";
            classer_path = startupPath + "\yolov8-detect-lable.txt";

            //Initialize the network class and read the local model
            opencv_net = CvDnn.ReadNetFromOnnx(model_path);

            result_array = new float[8400 * 56];
            factors = new float[2];

            result_pro = new PoseResult(factors);
        }

        private void button1_Click(object sender, EventArgs e)
        {
            OpenFileDialog ofd = new OpenFileDialog();
            ofd.Filter = fileFilter;
            if (ofd.ShowDialog() != DialogResult.OK) return;
            pictureBox1.Image = null;
            image_path = ofd.FileName;
            pictureBox1.Image = new Bitmap(image_path);
            textBox1.Text = "";
            image = new Mat(image_path);
            pictureBox2.Image = null;
        }

        private void button2_Click(object sender, EventArgs e)
        {
            if (image_path == "")
            {
                return;
            }

            //Zoom the picture
            max_image_length = image.Cols > image.Rows ? image.Cols : image.Rows;
            max_image = Mat.Zeros(new OpenCvSharp.Size(max_image_length, max_image_length), MatType.CV_8UC3);
            roi = new Rect(0, 0, image.Cols, image.Rows);
            image.CopyTo(new Mat(max_image, roi));

            factors[0] = factors[1] = (float)(max_image_length / 640.0);

            //Data normalization processing
            BN_image = CvDnn.BlobFromImage(max_image, 1 / 255.0, new OpenCvSharp.Size(640, 640), new Scalar(0, 0, 0), true, false);

            //Configure image input data
            opencv_net.SetInput(BN_image);

            dt1 = DateTime.Now;
            //Model inference, read inference results
            result_mat = opencv_net.Forward();
            dt2 = DateTime.Now;

            //Convert the inference result to float data type
            result_mat_to_float = new Mat(8400, 56, MatType.CV_32F, result_mat.Data);

            //Read data into array
            result_mat_to_float.GetArray<float>(out result_array);

            result = result_pro.process_result(result_array);

            result_image = result_pro.draw_result(result, image.Clone());

            if (!result_image.Empty())
            {
                pictureBox2.Image = new Bitmap(result_image.ToMemoryStream());
                sb.Clear();
                sb.AppendLine("Inference time consumption:" + (dt2 - dt1).TotalMilliseconds + "ms");
                sb.AppendLine("---------------------------------");
                textBox1.Text = sb.ToString();
            }
            else
            {
                textBox1.Text = "No information";
            }

        }
    }
}

Download

Demo download