MaskInverter<T>
Models & Types · AiDotNet.Diffusion.MaskUtilities
Inverts a mask so masked regions become unmasked and vice versa.
Implements: IDataTransformer<T, Tensor<T>, Tensor<T>>
For Beginners
If your mask highlights a person (white = person, black = background), inverting it highlights the background instead. This is useful when different tools use opposite conventions for what “masked” means.
How It Works
Mask inversion computes (1 - mask) for each pixel, flipping the masked and unmasked regions. This is commonly used when you want to inpaint the background instead of the foreground, or to convert between “keep” and “replace” mask conventions.
Constructors
| Constructor | Summary |
|---|---|
MaskInverter() |
Properties
| Property | Summary |
|---|---|
bool IsFitted { get; } |
|
bool SupportsInverseTransform { get; } |
|
int[] ColumnIndices { get; } |
Methods
| Method | Summary |
|---|---|
Tensor<T> Apply(Tensor<T> mask) |
Inverts a mask tensor by computing (1 - value) for each pixel. |
Tensor<T> FitTransform(Tensor<T> data) |
|
Tensor<T> InverseTransform(Tensor<T> data) |
|
Tensor<T> Transform(Tensor<T> data) |
|
string[] GetFeatureNamesOut(string[] inputFeatureNames) |
|
void Fit(Tensor<T> data) |