GaussianBlur

Kotlin |Java

@UnstableApi
public final class GaussianBlur extends SeparableConvolution

java.lang.Object
↳	androidx.media3.effect.SeparableConvolution
	↳	androidx.media3.effect.GaussianBlur

A SeparableConvolution to apply a Gaussian blur on image data.

The width of the blur is specified in pixels and applied symmetrically.

Summary

Public constructors
`GaussianBlur(float sigma)` Creates an instance with `numStandardDeviations` set to `2.0f`.
`GaussianBlur( @FloatRange(from = 0.0, fromInclusive = false) float sigma, @FloatRange(from = 0.0, fromInclusive = false) float numStandardDeviations )` Creates an instance.

Public methods
`ConvolutionFunction1D`	`getConvolution(long presentationTimeUs)` Returns a `1D convolution function`.

Inherited methods

From androidx.media3.common.Effect

long

getDurationAfterEffectApplied(long durationUs)

Returns the expected duration of the output stream when the effect is applied given a input durationUs.

From androidx.media3.effect.GlEffect

boolean

isNoOp(int inputWidth, int inputHeight)

Returns whether a GlEffect applies no change at every timestamp.

From androidx.media3.effect.SeparableConvolution

GlShaderProgram

toGlShaderProgram(Context context, boolean useHdr)

Returns a GlShaderProgram that applies the effect.

Public constructors

GaussianBlur

public GaussianBlur(float sigma)

Creates an instance with numStandardDeviations set to 2.0f.

Parameters
`float sigma`	The half-width of 1 standard deviation, in pixels.

GaussianBlur

public GaussianBlur(
    @FloatRange(from = 0.0, fromInclusive = false) float sigma,
    @FloatRange(from = 0.0, fromInclusive = false) float numStandardDeviations
)

Creates an instance.

Parameters
`@FloatRange(from = 0.0, fromInclusive = false) float sigma`	The half-width of 1 standard deviation, in pixels.
`@FloatRange(from = 0.0, fromInclusive = false) float numStandardDeviations`	The size of function domain, measured in the number of standard deviations.

Public methods

getConvolution

public ConvolutionFunction1D getConvolution(long presentationTimeUs)

Returns a 1D convolution function.

Parameters
`long presentationTimeUs`	The presentation timestamp of the input frame, in microseconds.