Skip to content

[Doc] Document image preprocessing steps for ONNX model input #255

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
yeelam-gordon wants to merge 1 commit into
base: docs
Choose a base branch
from
+12 −0
Open

[Doc] Document image preprocessing steps for ONNX model input #255

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 12 additions & 0 deletions docs/new-windows-ml/tutorial.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -273,6 +273,18 @@ Also see [Convert a model with AI Toolkit for VS Code](https://code.visualstudio

For API reference, see [**Ort::Session struct**](https://onnxruntime.ai/docs/api/c/struct_ort_1_1_session.html), [**Ort::MemoryInfo struct**](https://onnxruntime.ai/docs/api/c/struct_ort_1_1_memory_info.html), [**Ort::Value struct**](https://onnxruntime.ai/docs/api/c/struct_ort_1_1_value.html), [**Ort::AllocatorWithDefaultOptions struct**](https://onnxruntime.ai/docs/api/c/struct_ort_1_1_allocator_with_default_options.html), [**Ort::RunOptions struct**](https://onnxruntime.ai/docs/api/c/struct_ort_1_1_run_options.html).

### Image preprocessing

Image classification models like SqueezeNet expect input tensors in NCHW format (batch, channels, height, width) with normalized pixel values. The `PreprocessImageAsync` helper in the samples performs these steps:

1. **Load the image** as a `VideoFrame` using `VideoFrame.CreateWithSoftwareBitmap`
2. **Resize** to the model's expected dimensions (224×224 for SqueezeNet)
3. **Convert color format** from BGRA8 to RGB channels
4. **Normalize** pixel values using ImageNet mean `[0.485, 0.456, 0.406]` and standard deviation `[0.229, 0.224, 0.225]`
5. **Create a `DenseTensor<float>`** with shape `[1, 3, 224, 224]`

For the complete implementation, see [ImageProcessor.cs](https://github.com/microsoft/WindowsAppSDK-Samples/blob/main/Samples/WindowsML/Shared/cs/ImageProcessor.cs) in the Windows ML samples.

### [C#](#tab/csharp)

```csharp
Expand Down