Class MobileSAM

Predictor using MobileSAM models. Call SetImage(Texture) to specify the image you want to generate masks for. After setting an image you can call Predict(float[], float[]) to get the masks. The results will be stored in the Result RenderTexture.

Inheritance

object

MobileSAM

Implements

IDisposable

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Namespace: Doji.AI.Segmentation

Assembly: .dll

Syntax

public class MobileSAM : IDisposable

Constructors

MobileSAM()

Initializes a new instance of a mask predictor.

Declaration

public MobileSAM()

Properties

Backend

Which BackendType to run the model with.

Declaration

public BackendType Backend { get; set; }

Property Value

Type	Description
BackendType

Decoder

Declaration

public Model Decoder { get; }

Property Value

Type	Description
Model

Encoder

Declaration

public Model Encoder { get; }

Property Value

Type	Description
Model

OutputFormat

Which format the predicted masks are returned as.

Declaration

public RenderTextureFormat OutputFormat { get; set; }

Property Value

Type	Description
RenderTextureFormat

Result

Declaration

public RenderTexture Result { get; }

Property Value

Type	Description
RenderTexture

Methods

Dispose()

Performs application-defined tasks associated with freeing, releasing, or resetting unmanaged resources.

Declaration

public void Dispose()

Predict(float[], float[])

Predict masks for the given input prompts, using the currently set image.

Declaration

public void Predict(float[] pointCoords, float[] pointLabels)

Parameters

Type	Name	Description
float[]	pointCoords	A float array of point prompts to the model. Each point is given in pixel coordinates.
float[]	pointLabels	A length N array of labels for the point prompts. 1 indicates a foreground point and 0 indicates a background point.

SetImage(Tensor, (int height, int width))

Declaration

public void SetImage(Tensor transformedImage, (int height, int width) origSize)

Parameters

Type	Name	Description
Tensor	transformedImage
(int height, int width)	origSize

SetImage(Texture)

Specifies the image to be used for mask generation using the Predict(float[], float[]) method.

Declaration

public void SetImage(Texture image)

Parameters

Type	Name	Description
Texture	image

Remarks

This encodes the input image and stores the calculated image embeddings. The image only needs to be encoded once. Afterwards, you can query as many masks as you want.

Implements

IDisposable