Package org.opencv.video

Class DISOpticalFlow

java.lang.Object

org.opencv.core.Algorithm

org.opencv.video.DenseOpticalFlow

org.opencv.video.DISOpticalFlow

public class DISOpticalFlow
extends DenseOpticalFlow

DIS optical flow algorithm. This class implements the Dense Inverse Search (DIS) optical flow algorithm. More details about the algorithm can be found at CITE: Kroeger2016 . Includes three presets with preselected parameters to provide reasonable trade-off between speed and quality. However, even the slowest preset is still relatively fast, use DeepFlow if you need better quality and don't care about speed. This implementation includes several additional features compared to the algorithm described in the paper, including spatial propagation of flow vectors (REF: getUseSpatialPropagation), as well as an option to utilize an initial flow approximation passed to REF: calc (which is, essentially, temporal propagation, if the previous frame's flow field is passed).

Field Summary

Fields

Modifier and Type	Field	Description
static int	PRESET_FAST
static int	PRESET_MEDIUM
static int	PRESET_ULTRAFAST

Fields inherited from class org.opencv.core.Algorithm

nativeObj

Constructor Summary

Constructors

Modifier	Constructor	Description
protected	DISOpticalFlow(long addr)

Method Summary

Modifier and Type	Method	Description
static DISOpticalFlow	__fromPtr__(long addr)
static DISOpticalFlow	create()	Creates an instance of DISOpticalFlow
static DISOpticalFlow	create(int preset)	Creates an instance of DISOpticalFlow
protected void	finalize()
int	getFinestScale()	Finest level of the Gaussian pyramid on which the flow is computed (zero level corresponds to the original image resolution).
int	getGradientDescentIterations()	Maximum number of gradient descent iterations in the patch inverse search stage.
int	getPatchSize()	Size of an image patch for matching (in pixels).
int	getPatchStride()	Stride between neighbor patches.
boolean	getUseMeanNormalization()	Whether to use mean-normalization of patches when computing patch distance.
boolean	getUseSpatialPropagation()	Whether to use spatial propagation of good optical flow vectors.
float	getVariationalRefinementAlpha()	Weight of the smoothness term SEE: setVariationalRefinementAlpha
float	getVariationalRefinementDelta()	Weight of the color constancy term SEE: setVariationalRefinementDelta
float	getVariationalRefinementGamma()	Weight of the gradient constancy term SEE: setVariationalRefinementGamma
int	getVariationalRefinementIterations()	Number of fixed point iterations of variational refinement per scale.
void	setFinestScale(int val)	getFinestScale SEE: getFinestScale
void	setGradientDescentIterations(int val)	getGradientDescentIterations SEE: getGradientDescentIterations
void	setPatchSize(int val)	getPatchSize SEE: getPatchSize
void	setPatchStride(int val)	getPatchStride SEE: getPatchStride
void	setUseMeanNormalization(boolean val)	getUseMeanNormalization SEE: getUseMeanNormalization
void	setUseSpatialPropagation(boolean val)	getUseSpatialPropagation SEE: getUseSpatialPropagation
void	setVariationalRefinementAlpha(float val)	getVariationalRefinementAlpha SEE: getVariationalRefinementAlpha
void	setVariationalRefinementDelta(float val)	getVariationalRefinementDelta SEE: getVariationalRefinementDelta
void	setVariationalRefinementGamma(float val)	getVariationalRefinementGamma SEE: getVariationalRefinementGamma
void	setVariationalRefinementIterations(int val)	getGradientDescentIterations SEE: getGradientDescentIterations

Methods inherited from class org.opencv.video.DenseOpticalFlow

calc, collectGarbage

Methods inherited from class org.opencv.core.Algorithm

clear, empty, getDefaultName, getNativeObjAddr, save

Methods inherited from class java.lang.Object

clone, equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

PRESET_ULTRAFAST

public static final int PRESET_ULTRAFAST

See Also:

Constant Field Values

PRESET_FAST

public static final int PRESET_FAST

See Also:

Constant Field Values

PRESET_MEDIUM

public static final int PRESET_MEDIUM

See Also:

Constant Field Values

Constructor Details

DISOpticalFlow

protected DISOpticalFlow(long addr)

Method Details

__fromPtr__

public static DISOpticalFlow __fromPtr__(long addr)

getFinestScale

public int getFinestScale()

Finest level of the Gaussian pyramid on which the flow is computed (zero level corresponds to the original image resolution). The final flow is obtained by bilinear upscaling. SEE: setFinestScale

Returns:

automatically generated

setFinestScale

public void setFinestScale(int val)

getFinestScale SEE: getFinestScale

Parameters:

val - automatically generated

getPatchSize

public int getPatchSize()

Size of an image patch for matching (in pixels). Normally, default 8x8 patches work well enough in most cases. SEE: setPatchSize

Returns:

automatically generated

setPatchSize

public void setPatchSize(int val)

getPatchSize SEE: getPatchSize

Parameters:

val - automatically generated

getPatchStride

public int getPatchStride()

Stride between neighbor patches. Must be less than patch size. Lower values correspond to higher flow quality. SEE: setPatchStride

Returns:

automatically generated

setPatchStride

public void setPatchStride(int val)

getPatchStride SEE: getPatchStride

Parameters:

val - automatically generated

getGradientDescentIterations

public int getGradientDescentIterations()

Maximum number of gradient descent iterations in the patch inverse search stage. Higher values may improve quality in some cases. SEE: setGradientDescentIterations

Returns:

automatically generated

setGradientDescentIterations

public void setGradientDescentIterations(int val)

getGradientDescentIterations SEE: getGradientDescentIterations

Parameters:

val - automatically generated

getVariationalRefinementIterations

public int getVariationalRefinementIterations()

Number of fixed point iterations of variational refinement per scale. Set to zero to disable variational refinement completely. Higher values will typically result in more smooth and high-quality flow. SEE: setGradientDescentIterations

Returns:

automatically generated

setVariationalRefinementIterations

public void setVariationalRefinementIterations(int val)

getGradientDescentIterations SEE: getGradientDescentIterations

Parameters:

val - automatically generated

getVariationalRefinementAlpha

public float getVariationalRefinementAlpha()

Weight of the smoothness term SEE: setVariationalRefinementAlpha

Returns:

automatically generated

setVariationalRefinementAlpha

public void setVariationalRefinementAlpha(float val)

getVariationalRefinementAlpha SEE: getVariationalRefinementAlpha

Parameters:

val - automatically generated

getVariationalRefinementDelta

public float getVariationalRefinementDelta()

Weight of the color constancy term SEE: setVariationalRefinementDelta

Returns:

automatically generated

setVariationalRefinementDelta

public void setVariationalRefinementDelta(float val)

getVariationalRefinementDelta SEE: getVariationalRefinementDelta

Parameters:

val - automatically generated

getVariationalRefinementGamma

public float getVariationalRefinementGamma()

Weight of the gradient constancy term SEE: setVariationalRefinementGamma

Returns:

automatically generated

setVariationalRefinementGamma

public void setVariationalRefinementGamma(float val)

getVariationalRefinementGamma SEE: getVariationalRefinementGamma

Parameters:

val - automatically generated

getUseMeanNormalization

public boolean getUseMeanNormalization()

Whether to use mean-normalization of patches when computing patch distance. It is turned on by default as it typically provides a noticeable quality boost because of increased robustness to illumination variations. Turn it off if you are certain that your sequence doesn't contain any changes in illumination. SEE: setUseMeanNormalization

Returns:

automatically generated

setUseMeanNormalization

public void setUseMeanNormalization(boolean val)

getUseMeanNormalization SEE: getUseMeanNormalization

Parameters:

val - automatically generated

getUseSpatialPropagation

public boolean getUseSpatialPropagation()

Whether to use spatial propagation of good optical flow vectors. This option is turned on by default, as it tends to work better on average and can sometimes help recover from major errors introduced by the coarse-to-fine scheme employed by the DIS optical flow algorithm. Turning this option off can make the output flow field a bit smoother, however. SEE: setUseSpatialPropagation

Returns:

automatically generated

setUseSpatialPropagation

public void setUseSpatialPropagation(boolean val)

getUseSpatialPropagation SEE: getUseSpatialPropagation

Parameters:

val - automatically generated

create

public static DISOpticalFlow create(int preset)

Creates an instance of DISOpticalFlow

Parameters:

preset - one of PRESET_ULTRAFAST, PRESET_FAST and PRESET_MEDIUM

Returns:

automatically generated

create

public static DISOpticalFlow create()

Creates an instance of DISOpticalFlow

Returns:

automatically generated

finalize

protected void finalize() throws Throwable

Overrides:

finalize in class DenseOpticalFlow

Throws:

Throwable