YoloDotNet

YoloDotNet is a C# .NET 8.0 implementation of Yolov8 and ONNX runtime with CUDA

YoloDotNet is a .NET 8 implementation of Yolov8 for detecting objects in images and videos using ML.NET and ONNX runtime with GPU acceleration using CUDA. YoloDotNet currently supports Classification, Object Detection and Segmentation in both images and videos.

Classification	Object Dectection	Segmentation
Categorize an image	Detect multiple objects in a single image	Separate detected objects by pixel maps
image from pexels.com	image from pexels.com	image from pexels.com

YoloDotNet with GPU-acceleration requires CUDA and cuDNN.

ℹ️ Before installing CUDA and cuDNN, make sure to verify the ONNX runtime's current compatibility with specific versions.

Download and install CUDA
Download cuDNN and follow the installation instructions

Yolov8 model exported to ONNX format
Currently, YoloDotNet supports Classification, Object Detection and Segmentation in both images and videos

Verify your model

using YoloDotNet;

// Instantiate a new Yolo object with your ONNX-model
using var yolo = new Yolo(@"path\to\model.onnx");

Console.WriteLine(yolo.OnnxModel.ModelType); // Output if valid: Classification or ObjectDetection

Note

For Video, you need FFmpeg and FFProbe

Download FFMPEG
Add FFmpeg and ffprobe to the Path-variable in your Environment Variables

Nuget

> dotnet add package YoloDotNet

Example - Image

using SixLabors.ImageSharp;
using SixLabors.ImageSharp.PixelFormats;
using YoloDotNet;
using YoloDotNet.Extensions;

// Instantiate a new Yolo object with your ONNX-model and CUDA (default)
using var yolo = new Yolo(@"path\to\your_model.onnx");

// Load image
using var image = Image.Load<Rgba32>(@"path\to\image.jpg");

// Run
var results = yolo.RunClassification(image, 5); // Top 5 classes
//var results = yolo.RunObjectDetection(image, 0.25);
//var results = yolo.RunSegmentation(image, 0.25);

image.Draw(results);
image.Save(@"path\to\save\image.jpg");

Example - Video

using SixLabors.ImageSharp;
using SixLabors.ImageSharp.PixelFormats;
using YoloDotNet;
using YoloDotNet.Extensions;

// Instantiate a new Yolo object with your ONNX-model and CUDA
using var yolo = new Yolo(@"path\to\your_model.onnx");

// Video options
var options = new VideoOptions
{
    VideoFile = @"path\to\video.mp4",
    OutputDir = @"path\to\output\folder",
    //GenerateVideo = true,
    //DrawLabels = true,
    //FPS = 30,
    //Width = 1280,
    //Height = 720,
    //DrawConfidence = true,
    //KeepAudio = true,
    //KeepFrames = false
};

// Run
var results = yolo.RunClassification(options, 5); // Top 5 classes
//var results = yolo.RunObjectDetection(options, 0.25);
//var results = yolo.RunSegmentation(options, 0.25);

// Do further processing with results if needed...

GPU

Object detection with GPU and GPU-Id = 0 is enabled by default

// Default setup. GPU with GPU-Id 0
using var yolo = new Yolo(@"path\to\model.onnx");

With a specific GPU-Id

// GPU with a user defined GPU-Id
using var yolo = new Yolo(@"path\to\model.onnx", true, 1);

CPU

YoloDotNet detection with CPU

// With CPU
using var yolo = new Yolo(@"path\to\model.onnx", false);

Access ONNX metadata and labels

The internal ONNX metadata such as input & output parameters, version, author, description, date along with the labels can be accessed via the yolo.OnnxModel property.

Example:

using var yolo = new Yolo(@"path\to\model.onnx");

// ONNX metadata and labels resides inside yolo.OnnxModel
Console.WriteLine(yolo.OnnxModel);

Example:

// Instantiate a new object
using var yolo = new Yolo(@"path\to\model.onnx");

// Display metadata
foreach (var property in yolo.OnnxModel.GetType().GetProperties())
{
    var value = property.GetValue(yolo.OnnxModel);
    Console.WriteLine($"{property.Name,-20}{value!}");

    if (property.Name == nameof(yolo.OnnxModel.CustomMetaData))
        foreach (var data in (Dictionary<string, string>)value!)
            Console.WriteLine($"{"",-20}{data.Key,-20}{data.Value}");
}

// Get ONNX labels
var labels = yolo.OnnxModel.Labels;

Console.WriteLine();
Console.WriteLine($"Labels ({labels.Length}):");
Console.WriteLine(new string('-', 58));

// Display
for (var i = 0; i < labels.Length; i++)
    Console.WriteLine($"index: {i,-8} label: {labels[i].Name,20} color: {labels[i].Color}");

// Output:

// ModelType           ObjectDetection
// InputName           images
// OutputName          output0
// CustomMetaData      System.Collections.Generic.Dictionary`2[System.String,System.String]
//                     date                2023-11-07T13:33:33.565196
//                     description         Ultralytics YOLOv8n model trained on coco.yaml
//                     author              Ultralytics
//                     task                detect
//                     license             AGPL-3.0 https://ultralytics.com/license
//                     version             8.0.202
//                     stride              32
//                     batch               1
//                     imgsz               [640, 640]
//                     names               {0: 'person', 1: 'bicycle', 2: 'car' ... }
// ImageSize           Size [ Width=640, Height=640 ]
// Input               Input { BatchSize = 1, Channels = 3, Width = 640, Height = 640 }
// Output              ObjectDetectionShape { BatchSize = 1, Elements = 84, Channels = 8400 }
// Labels              YoloDotNet.Models.LabelModel[]
//
// Labels (80):
// ---------------------------------------------------------
// index: 0        label: person              color: #5d8aa8
// index: 1        label: bicycle             color: #f0f8ff
// index: 2        label: car                 color: #e32636
// index: 3        label: motorcycle          color: #efdecd
// ...

Donate

https://paypal.me/nickswardh

References & Acknowledgements

https://github.com/ultralytics/ultralytics

https://github.com/sstainba/Yolov8.Net

https://github.com/mentalstack/yolov5-net

Steven-L-42/YoloDotNet