SemanticSegPreprocessor<T>
Models & Types ยท AiDotNet.Diffusion.Preprocessing
Semantic segmentation preprocessor for ControlNet conditioning.
Inheritance: DiffusionPreprocessorBase<T> โ SemanticSegPreprocessor<T>
For Beginners
This labels every pixel in the image (sky, person, car, etc.) and paints them different colors. ControlNet uses this to generate images where objects are in the same regions.
How It Works
Produces semantic segmentation maps where each pixel is assigned a class label encoded as a color. The output guides ControlNet to respect object boundaries and regions.
Constructors
| Constructor | Summary |
|---|---|
SemanticSegPreprocessor() |
Properties
| Property | Summary |
|---|---|
ControlType OutputControlType { get; } |
|
int OutputChannels { get; } |
Methods
| Method | Summary |
|---|---|
Tensor<T> Transform(Tensor<T> data) |