OpenCV  4.5.1 Open Source Computer Vision
cv::DISOpticalFlow Class Referenceabstract

DIS optical flow algorithm. More...

#include <opencv2/video/tracking.hpp>

Inheritance diagram for cv::DISOpticalFlow:

## Public Types

enum  {
PRESET_ULTRAFAST = 0,
PRESET_FAST = 1,
PRESET_MEDIUM = 2
}

## Public Member Functions

virtual int getFinestScale () const =0
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. More...

virtual int getGradientDescentIterations () const =0
Maximum number of gradient descent iterations in the patch inverse search stage. Higher values may improve quality in some cases. More...

virtual int getPatchSize () const =0
Size of an image patch for matching (in pixels). Normally, default 8x8 patches work well enough in most cases. More...

virtual int getPatchStride () const =0
Stride between neighbor patches. Must be less than patch size. Lower values correspond to higher flow quality. More...

virtual bool getUseMeanNormalization () const =0
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. More...

virtual bool getUseSpatialPropagation () const =0
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. More...

virtual float getVariationalRefinementAlpha () const =0
Weight of the smoothness term. More...

virtual float getVariationalRefinementDelta () const =0
Weight of the color constancy term. More...

virtual float getVariationalRefinementGamma () const =0
Weight of the gradient constancy term. More...

virtual int getVariationalRefinementIterations () const =0
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. More...

virtual void setFinestScale (int val)=0
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. More...

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

virtual void setPatchSize (int val)=0
Size of an image patch for matching (in pixels). Normally, default 8x8 patches work well enough in most cases. More...

virtual void setPatchStride (int val)=0
Stride between neighbor patches. Must be less than patch size. Lower values correspond to higher flow quality. More...

virtual void setUseMeanNormalization (bool val)=0
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. More...

virtual void setUseSpatialPropagation (bool val)=0
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. More...

virtual void setVariationalRefinementAlpha (float val)=0
Weight of the smoothness term. More...

virtual void setVariationalRefinementDelta (float val)=0
Weight of the color constancy term. More...

virtual void setVariationalRefinementGamma (float val)=0
Weight of the gradient constancy term. More...

virtual void setVariationalRefinementIterations (int val)=0
Maximum number of gradient descent iterations in the patch inverse search stage. Higher values may improve quality in some cases. More...

Public Member Functions inherited from cv::DenseOpticalFlow
virtual void calc (InputArray I0, InputArray I1, InputOutputArray flow)=0
Calculates an optical flow. More...

virtual void collectGarbage ()=0
Releases all inner buffers. More...

Public Member Functions inherited from cv::Algorithm
Algorithm ()

virtual ~Algorithm ()

virtual void clear ()
Clears the algorithm state. More...

virtual bool empty () const
Returns true if the Algorithm is empty (e.g. in the very beginning or after unsuccessful read. More...

virtual String getDefaultName () const

virtual void read (const FileNode &fn)
Reads algorithm parameters from a file storage. More...

virtual void save (const String &filename) const

virtual void write (FileStorage &fs) const
Stores algorithm parameters in a file storage. More...

void write (const Ptr< FileStorage > &fs, const String &name=String()) const
simplified API for language bindings This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts. More...

## Static Public Member Functions

static Ptr< DISOpticalFlowcreate (int preset=DISOpticalFlow::PRESET_FAST)
Creates an instance of DISOpticalFlow. More...

Static Public Member Functions inherited from cv::Algorithm
template<typename _Tp >
static Ptr< _Tp > load (const String &filename, const String &objname=String())
Loads algorithm from the file. More...

template<typename _Tp >
static Ptr< _Tp > loadFromString (const String &strModel, const String &objname=String())
Loads algorithm from a String. More...

template<typename _Tp >
static Ptr< _Tp > read (const FileNode &fn)
Reads algorithm from the file node. More...

Protected Member Functions inherited from cv::Algorithm
void writeFormat (FileStorage &fs) const

## Detailed Description

DIS optical flow algorithm.

This class implements the Dense Inverse Search (DIS) optical flow algorithm. More details about the algorithm can be found at [126] . 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 (getUseSpatialPropagation), as well as an option to utilize an initial flow approximation passed to calc (which is, essentially, temporal propagation, if the previous frame's flow field is passed).

## ◆ anonymous enum

 anonymous enum
Enumerator
PRESET_ULTRAFAST
PRESET_FAST
PRESET_MEDIUM

## ◆ create()

 static Ptr cv::DISOpticalFlow::create ( int preset = DISOpticalFlow::PRESET_FAST )
static
Python:
retval=cv.DISOpticalFlow_create([, preset])

Creates an instance of DISOpticalFlow.

Parameters
 preset one of PRESET_ULTRAFAST, PRESET_FAST and PRESET_MEDIUM

## ◆ getFinestScale()

 virtual int cv::DISOpticalFlow::getFinestScale ( ) const
pure virtual
Python:
retval=cv.DISOpticalFlow.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.

setFinestScale

 virtual int cv::DISOpticalFlow::getGradientDescentIterations ( ) const
pure virtual
Python:

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

## ◆ getPatchSize()

 virtual int cv::DISOpticalFlow::getPatchSize ( ) const
pure virtual
Python:
retval=cv.DISOpticalFlow.getPatchSize()

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

setPatchSize

## ◆ getPatchStride()

 virtual int cv::DISOpticalFlow::getPatchStride ( ) const
pure virtual
Python:
retval=cv.DISOpticalFlow.getPatchStride()

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

setPatchStride

## ◆ getUseMeanNormalization()

 virtual bool cv::DISOpticalFlow::getUseMeanNormalization ( ) const
pure virtual
Python:
retval=cv.DISOpticalFlow.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.

setUseMeanNormalization

## ◆ getUseSpatialPropagation()

 virtual bool cv::DISOpticalFlow::getUseSpatialPropagation ( ) const
pure virtual
Python:
retval=cv.DISOpticalFlow.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.

setUseSpatialPropagation

## ◆ getVariationalRefinementAlpha()

 virtual float cv::DISOpticalFlow::getVariationalRefinementAlpha ( ) const
pure virtual
Python:
retval=cv.DISOpticalFlow.getVariationalRefinementAlpha()

Weight of the smoothness term.

setVariationalRefinementAlpha

## ◆ getVariationalRefinementDelta()

 virtual float cv::DISOpticalFlow::getVariationalRefinementDelta ( ) const
pure virtual
Python:
retval=cv.DISOpticalFlow.getVariationalRefinementDelta()

Weight of the color constancy term.

setVariationalRefinementDelta

## ◆ getVariationalRefinementGamma()

 virtual float cv::DISOpticalFlow::getVariationalRefinementGamma ( ) const
pure virtual
Python:
retval=cv.DISOpticalFlow.getVariationalRefinementGamma()

Weight of the gradient constancy term.

setVariationalRefinementGamma

## ◆ getVariationalRefinementIterations()

 virtual int cv::DISOpticalFlow::getVariationalRefinementIterations ( ) const
pure virtual
Python:
retval=cv.DISOpticalFlow.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.

## ◆ setFinestScale()

 virtual void cv::DISOpticalFlow::setFinestScale ( int val )
pure virtual
Python:
None=cv.DISOpticalFlow.setFinestScale(val)

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.

getFinestScale

 virtual void cv::DISOpticalFlow::setGradientDescentIterations ( int val )
pure virtual
Python:

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

## ◆ setPatchSize()

 virtual void cv::DISOpticalFlow::setPatchSize ( int val )
pure virtual
Python:
None=cv.DISOpticalFlow.setPatchSize(val)

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

getPatchSize

## ◆ setPatchStride()

 virtual void cv::DISOpticalFlow::setPatchStride ( int val )
pure virtual
Python:
None=cv.DISOpticalFlow.setPatchStride(val)

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

getPatchStride

## ◆ setUseMeanNormalization()

 virtual void cv::DISOpticalFlow::setUseMeanNormalization ( bool val )
pure virtual
Python:
None=cv.DISOpticalFlow.setUseMeanNormalization(val)

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.

getUseMeanNormalization

## ◆ setUseSpatialPropagation()

 virtual void cv::DISOpticalFlow::setUseSpatialPropagation ( bool val )
pure virtual
Python:
None=cv.DISOpticalFlow.setUseSpatialPropagation(val)

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.

getUseSpatialPropagation

## ◆ setVariationalRefinementAlpha()

 virtual void cv::DISOpticalFlow::setVariationalRefinementAlpha ( float val )
pure virtual
Python:
None=cv.DISOpticalFlow.setVariationalRefinementAlpha(val)

Weight of the smoothness term.

getVariationalRefinementAlpha

## ◆ setVariationalRefinementDelta()

 virtual void cv::DISOpticalFlow::setVariationalRefinementDelta ( float val )
pure virtual
Python:
None=cv.DISOpticalFlow.setVariationalRefinementDelta(val)

Weight of the color constancy term.

getVariationalRefinementDelta

## ◆ setVariationalRefinementGamma()

 virtual void cv::DISOpticalFlow::setVariationalRefinementGamma ( float val )
pure virtual
Python:
None=cv.DISOpticalFlow.setVariationalRefinementGamma(val)

Weight of the gradient constancy term.

getVariationalRefinementGamma

## ◆ setVariationalRefinementIterations()

 virtual void cv::DISOpticalFlow::setVariationalRefinementIterations ( int val )
pure virtual
Python:
None=cv.DISOpticalFlow.setVariationalRefinementIterations(val)

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