Contoso.AI.ImageSegmenterSINet 0.1.6-beta

.NET 8.0 This package targets .NET 8.0. The package is compatible with this framework or higher.
This is a prerelease version of Contoso.AI.ImageSegmenterSINet. 
dotnet add package Contoso.AI.ImageSegmenterSINet --version 0.1.6-beta
                    


                    

                
NuGet\Install-Package Contoso.AI.ImageSegmenterSINet -Version 0.1.6-beta
This command is intended to be used within the Package Manager Console in Visual Studio, as it uses the NuGet module's version of Install-Package. 
<PackageReference Include="Contoso.AI.ImageSegmenterSINet" Version="0.1.6-beta" />
For projects that support PackageReference, copy this XML node into the project file to reference the package. 
<PackageVersion Include="Contoso.AI.ImageSegmenterSINet" Version="0.1.6-beta" />
                    


                            Directory.Packages.props
                        

                    

                
<PackageReference Include="Contoso.AI.ImageSegmenterSINet" />
                    


                            Project file
For projects that support Central Package Management (CPM), copy this XML node into the solution Directory.Packages.props file to version the package. 
paket add Contoso.AI.ImageSegmenterSINet --version 0.1.6-beta
                    


                    

                
#r "nuget: Contoso.AI.ImageSegmenterSINet, 0.1.6-beta"
#r directive can be used in F# Interactive and Polyglot Notebooks. Copy this into the interactive tool or source code of the script to reference the package. 
#:package Contoso.AI.ImageSegmenterSINet@0.1.6-beta
#:package directive can be used in C# file-based apps starting in .NET 10 preview 4. Copy this into a .cs file before any lines of code to reference the package. 
#addin nuget:?package=Contoso.AI.ImageSegmenterSINet&version=0.1.6-beta&prerelease
                    


                            Install as a Cake Addin
                        

                    

                
#tool nuget:?package=Contoso.AI.ImageSegmenterSINet&version=0.1.6-beta&prerelease
                    


                            Install as a Cake Tool

Contoso.AI.ImageSegmenterSINet

AI-powered image segmentation using Qualcomm's SINet (Salient Image Network) ONNX model. Separates foreground from background in images with NPU hardware acceleration.

Quick Start

using Contoso.AI;
using System.Drawing;

// Check if the feature is ready
var readyState = ImageSegmenterSINet.GetReadyState();

if (readyState != AIFeatureReadyState.Ready)
{
    // Prepare the feature (downloads QNN Execution Provider if needed)
    var readyResult = await ImageSegmenterSINet.EnsureReadyAsync();
    if (readyResult.Status != AIFeatureReadyResultState.Success)
    {
        Console.WriteLine($"Failed to initialize: {readyResult.ExtendedError?.Message}");
        return;
    }
}

// Create segmenter instance
using var segmenter = await ImageSegmenterSINet.CreateAsync();

// Segment an image
using var bitmap = new Bitmap("photo.jpg");
var result = segmenter.SegmentImage(bitmap);

// Extract foreground with transparent background
using var foreground = result.ExtractForeground(bitmap);
foreground.Save("foreground.png");

Features

  • ✅ Foreground/background segmentation using SINet
  • ✅ NPU hardware acceleration (QNN Execution Provider)
  • ✅ Automatic model download at build time (both float and quantized models)
  • ✅ Intelligent model selection: quantized for NPU, float for CPU
  • ✅ Easy-to-use async factory pattern
  • ✅ Multiple output formats (mask, overlay, extracted foreground)

Requirements

  • Windows 10 SDK 19041+
  • .NET 8.0+
  • NPU recommended for best performance (falls back to CPU automatically)

Model Selection

The library automatically downloads two ONNX models at build time:

  • Quantized (int8) model - Optimized for QNN NPU acceleration
  • Float (fp32) model - Optimized for CPU execution

At runtime, CreateAsync() intelligently selects:

  • Quantized model when QNN NPU is available
  • Float model when falling back to CPU

This ensures optimal performance and compatibility across different hardware configurations.

API

Method Description
ImageSegmenterSINet.GetReadyState() Check if feature is available
ImageSegmenterSINet.EnsureReadyAsync() Prepare dependencies
ImageSegmenterSINet.CreateAsync() Create segmenter instance
SegmentImage(Bitmap) Perform segmentation
result.CreateMaskOverlay(Bitmap, Color?) Visualize segmentation on original image
result.ExtractForeground(Bitmap) Extract foreground with transparency

License

MIT License - see LICENSE file for details.

Product Compatible and additional computed target framework versions.
.NET net8.0-windows10.0.19041 is compatible.  net9.0-windows was computed.  net10.0-windows was computed. 
Compatible target framework(s)
Included target framework(s) (in package)
Learn more about Target Frameworks and .NET Standard.

NuGet packages

This package is not used by any NuGet packages.

GitHub repositories

This package is not used by any popular GitHub repositories.

Version Downloads Last Updated
0.1.6-beta 46 2/5/2026
0.1.1-beta 46 2/4/2026
0.1.0-beta 42 2/4/2026