Skip to content

cs-wpf

Directory actions

More options

Directory actions

More options

Latest commit

 

History

History

cs-wpf

README.md

Windows ML WPF Image Classification Sample

This sample demonstrates how to use Windows ML in a WPF (Windows Presentation Foundation) desktop application for real-time image classification. The application provides a user-friendly interface for selecting images, configuring execution providers, and viewing classification results.

Features

  • Image File Selection - Browse and select images for classification using standard file picker
  • Execution Provider Configuration - View available providers (NPU, GPU, CPU) and configure download settings
  • Real-time Inference - Classify images using the SqueezeNet model with immediate results
  • Results Display - Show top 5 predictions with confidence percentages in formatted output
  • Provider Information - Display detailed execution provider status and capabilities

Key Technologies

  • WPF - Windows Presentation Foundation for desktop UI
  • Windows ML - ONNX Runtime integration via Windows App SDK
  • Shared Helpers - Uses WindowsML.Shared library for consistent ML operations

Building and Running

  1. Open the main solution: WindowsML-Samples.sln
  2. Set the WindowsMLSampleForWPF project as the startup project
  3. Build and run (F5)

Usage

  1. Launch Application - The app loads and displays available execution providers
  2. Configure Providers - Optionally check "Allow Provider Download" to enable automatic EP downloads
  3. Select Image - Click "Select Image" to browse for an image file (.jpg, .jpeg, .png)
  4. Run Classification - Click "Run Inference" to classify the selected image
  5. View Results - See the top 5 predictions with confidence percentages

Expected Output


Top 5 Predictions:
-------------------------------------------
Label                            Confidence
-------------------------------------------
French bulldog                       45.07%
bull mastiff                         35.05%
boxer                                 2.85%
pug                                   1.01%
American Staffordshire terrier        0.57%
-------------------------------------------

Code Structure

The sample uses the shared WindowsML.Shared library for consistency across samples:

// Initialize execution providers
var catalog = ExecutionProviderCatalog.GetDefault();
await ModelManager.InitializeExecutionProvidersAsync(allowDownload: allowDownload);

// Load and preprocess image
var videoFrame = await ImageProcessor.LoadImageFileAsync(_selectedImagePath);
var inputTensor = await ImageProcessor.PreprocessImageAsync(videoFrame);

// Run inference
using var results = InferenceEngine.RunInference(_session, inputTensor);
var resultTensor = InferenceEngine.ExtractResults(_session, results);

// Process results
var topPredictions = ResultProcessor.GetTopPredictions(resultTensor, _labels, 5);

Related Samples

Model Information

This sample uses the SqueezeNet model:

  • Purpose: Image classification with 1000 categories
  • Input: 224x224 RGB images
  • Output: Probability distribution over ImageNet classes
  • Size: Lightweight model optimized for performance