Fast dense optical flow computation based on robust local optical flow (RLOF) algorithms and sparse-to-dense interpolation scheme. More...

#include <opencv2/optflow/rlofflow.hpp>

Collaboration diagram for cv::optflow::DenseRLOFOpticalFlow:

Public Member Functions
virtual int	getEPICK () const =0
	see ximgproc::EdgeAwareInterpolator() K value.

virtual float	getEPICLambda () const =0
	see ximgproc::EdgeAwareInterpolator() lambda value.

virtual float	getEPICSigma () const =0
	see ximgproc::EdgeAwareInterpolator() sigma value.

virtual float	getFgsLambda () const =0
	see ximgproc::EdgeAwareInterpolator().

virtual float	getFgsSigma () const =0
	see ximgproc::EdgeAwareInterpolator().

virtual float	getForwardBackward () const =0
	Threshold for the forward backward confidence check.

virtual Size	getGridStep () const =0
	Size of the grid to spawn the motion vectors.

virtual InterpolationType	getInterpolation () const =0
	Interpolation used to compute the dense optical flow.

virtual int	getRICSLICType () const =0
	Parameter to choose superpixel algorithm variant to use:

virtual int	getRICSPSize () const =0
	Parameter to tune the approximate size of the superpixel used for oversegmentation.

virtual Ptr< RLOFOpticalFlowParameter >	getRLOFOpticalFlowParameter () const =0
	Configuration of the RLOF alogrithm.

virtual bool	getUsePostProc () const =0
	enables ximgproc::fastGlobalSmootherFilter

virtual bool	getUseVariationalRefinement () const =0
	enables VariationalRefinement

virtual void	setEPICK (int val)=0
	see ximgproc::EdgeAwareInterpolator() K value.

virtual void	setEPICLambda (float val)=0
	see ximgproc::EdgeAwareInterpolator() lambda value.

virtual void	setEPICSigma (float val)=0
	see ximgproc::EdgeAwareInterpolator() sigma value.

virtual void	setFgsLambda (float val)=0
	see ximgproc::EdgeAwareInterpolator().

virtual void	setFgsSigma (float val)=0
	see ximgproc::EdgeAwareInterpolator().

virtual void	setForwardBackward (float val)=0
	Threshold for the forward backward confidence check.

virtual void	setGridStep (Size val)=0
	Size of the grid to spawn the motion vectors.

virtual void	setInterpolation (InterpolationType val)=0
	Interpolation used to compute the dense optical flow.

virtual void	setRICSLICType (int val)=0
	Parameter to choose superpixel algorithm variant to use:

virtual void	setRICSPSize (int val)=0
	Parameter to tune the approximate size of the superpixel used for oversegmentation.

virtual void	setRLOFOpticalFlowParameter (Ptr< RLOFOpticalFlowParameter > val)=0
	Configuration of the RLOF alogrithm.

virtual void	setUsePostProc (bool val)=0
	enables ximgproc::fastGlobalSmootherFilter

virtual void	setUseVariationalRefinement (bool val)=0
	enables VariationalRefinement

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

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

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

virtual	~Algorithm ()

virtual void	clear ()
	Clears the algorithm state.

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

virtual String	getDefaultName () const

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

virtual void	save (const String &filename) const

void	write (const Ptr< FileStorage > &fs, const String &name=String()) const

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

void	write (FileStorage &fs, const String &name) const

Static Public Member Functions
static Ptr< DenseRLOFOpticalFlow >	create (Ptr< RLOFOpticalFlowParameter > rlofParam=Ptr< RLOFOpticalFlowParameter >(), float forwardBackwardThreshold=1.f, Size gridStep=Size(6, 6), InterpolationType interp_type=InterpolationType::INTERP_EPIC, int epicK=128, float epicSigma=0.05f, float epicLambda=999.0f, int ricSPSize=15, int ricSLICType=100, bool use_post_proc=true, float fgsLambda=500.0f, float fgsSigma=1.5f, bool use_variational_refinement=false)
	Creates instance of optflow::DenseRLOFOpticalFlow.

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.

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

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

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

Detailed Description

Fast dense optical flow computation based on robust local optical flow (RLOF) algorithms and sparse-to-dense interpolation scheme.

The RLOF is a fast local optical flow approach described in [238] [239] [240] and [241] similar to the pyramidal iterative Lucas-Kanade method as proposed by [36]. More details and experiments can be found in the following thesis [242]. The implementation is derived from optflow::calcOpticalFlowPyrLK().

The sparse-to-dense interpolation scheme allows for fast computation of dense optical flow using RLOF (see [103]). For this scheme the following steps are applied:

motion vector seeded at a regular sampled grid are computed. The sparsity of this grid can be configured with setGridStep
(optinally) errornous motion vectors are filter based on the forward backward confidence. The threshold can be configured with setForwardBackward. The filter is only applied if the threshold >0 but than the runtime is doubled due to the estimation of the backward flow.
Vector field interpolation is applied to the motion vector set to obtain a dense vector field.

For the RLOF configuration see optflow::RLOFOpticalFlowParameter for further details. Parameters have been described in [238] [239] [240] and [241].

Note: If the grid size is set to (1,1) and the forward backward threshold <= 0 than pixelwise dense optical flow field is computed by RLOF without using interpolation.; Note that in output, if no correspondences are found between I0 and I1, the flow is set to 0.

See also: optflow::calcOpticalFlowDenseRLOF(), optflow::RLOFOpticalFlowParameter

Member Function Documentation

◆ create()

static Ptr< DenseRLOFOpticalFlow > cv::optflow::DenseRLOFOpticalFlow::create	(	Ptr< RLOFOpticalFlowParameter >	rlofParam = Ptr< RLOFOpticalFlowParameter >(),
		float	forwardBackwardThreshold = 1.f,
		Size	gridStep = Size(6, 6),
		InterpolationType	interp_type = InterpolationType::INTERP_EPIC,
		int	epicK = 128,
		float	epicSigma = 0.05f,
		float	epicLambda = 999.0f,
		int	ricSPSize = 15,
		int	ricSLICType = 100,
		bool	use_post_proc = true,
		float	fgsLambda = 500.0f,
		float	fgsSigma = 1.5f,
		bool	use_variational_refinement = false )

static

Python:
	cv.optflow.DenseRLOFOpticalFlow.create(	[, rlofParam[, forwardBackwardThreshold[, gridStep[, interp_type[, epicK[, epicSigma[, epicLambda[, ricSPSize[, ricSLICType[, use_post_proc[, fgsLambda[, fgsSigma[, use_variational_refinement]]]]]]]]]]]]]	) ->	retval
	cv.optflow.DenseRLOFOpticalFlow_create(	[, rlofParam[, forwardBackwardThreshold[, gridStep[, interp_type[, epicK[, epicSigma[, epicLambda[, ricSPSize[, ricSLICType[, use_post_proc[, fgsLambda[, fgsSigma[, use_variational_refinement]]]]]]]]]]]]]	) ->	retval

Creates instance of optflow::DenseRLOFOpticalFlow.

Parameters

rlofParam	see optflow::RLOFOpticalFlowParameter
forwardBackwardThreshold	see setForwardBackward
gridStep	see setGridStep
interp_type	see setInterpolation
epicK	see setEPICK
epicSigma	see setEPICSigma
epicLambda	see setEPICLambda
ricSPSize	see setRICSPSize
ricSLICType	see setRICSLICType
use_post_proc	see setUsePostProc
fgsLambda	see setFgsLambda
fgsSigma	see setFgsSigma
use_variational_refinement	see setUseVariationalRefinement

◆ getEPICK()

virtual int cv::optflow::DenseRLOFOpticalFlow::getEPICK ( ) const

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.getEPICK(		) ->	retval

see ximgproc::EdgeAwareInterpolator() K value.

K is a number of nearest-neighbor matches considered, when fitting a locally affine model. Usually it should be around 128. However, lower values would make the interpolation noticeably faster.

See also: ximgproc::EdgeAwareInterpolator, setEPICK

◆ getEPICLambda()

virtual float cv::optflow::DenseRLOFOpticalFlow::getEPICLambda ( ) const

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.getEPICLambda(		) ->	retval

see ximgproc::EdgeAwareInterpolator() lambda value.

Lambda is a parameter defining the weight of the edge-aware term in geodesic distance, should be in the range of 0 to 1000.

See also: ximgproc::EdgeAwareInterpolator, setEPICSigma

◆ getEPICSigma()

virtual float cv::optflow::DenseRLOFOpticalFlow::getEPICSigma ( ) const

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.getEPICSigma(		) ->	retval

see ximgproc::EdgeAwareInterpolator() sigma value.

Sigma is a parameter defining how fast the weights decrease in the locally-weighted affine fitting. Higher values can help preserve fine details, lower values can help to get rid of noise in the output flow.

See also: ximgproc::EdgeAwareInterpolator, setEPICSigma

◆ getFgsLambda()

virtual float cv::optflow::DenseRLOFOpticalFlow::getFgsLambda ( ) const

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.getFgsLambda(		) ->	retval

see ximgproc::EdgeAwareInterpolator().

Sets the respective fastGlobalSmootherFilter() parameter.

See also: ximgproc::EdgeAwareInterpolator, setFgsLambda

◆ getFgsSigma()

virtual float cv::optflow::DenseRLOFOpticalFlow::getFgsSigma ( ) const

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.getFgsSigma(		) ->	retval

see ximgproc::EdgeAwareInterpolator().

Sets the respective fastGlobalSmootherFilter() parameter.

See also: ximgproc::EdgeAwareInterpolator, ximgproc::fastGlobalSmootherFilter, setFgsSigma

◆ getForwardBackward()

virtual float cv::optflow::DenseRLOFOpticalFlow::getForwardBackward ( ) const

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.getForwardBackward(		) ->	retval

Threshold for the forward backward confidence check.

See also: setForwardBackward

◆ getGridStep()

virtual Size cv::optflow::DenseRLOFOpticalFlow::getGridStep ( ) const

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.getGridStep(		) ->	retval

Size of the grid to spawn the motion vectors.

For each grid point a motion vector is computed. Some motion vectors will be removed due to the forwatd backward threshold (if set >0). The rest will be the base of the vector field interpolation.

See also: getForwardBackward, setGridStep

◆ getInterpolation()

virtual InterpolationType cv::optflow::DenseRLOFOpticalFlow::getInterpolation ( ) const

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.getInterpolation(		) ->	retval

Interpolation used to compute the dense optical flow.

See also: ximgproc::EdgeAwareInterpolator, setInterpolation

◆ getRICSLICType()

virtual int cv::optflow::DenseRLOFOpticalFlow::getRICSLICType ( ) const

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.getRICSLICType(		) ->	retval

Parameter to choose superpixel algorithm variant to use:

See also: setRICSLICType

◆ getRICSPSize()

virtual int cv::optflow::DenseRLOFOpticalFlow::getRICSPSize ( ) const

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.getRICSPSize(		) ->	retval

Parameter to tune the approximate size of the superpixel used for oversegmentation.

See also: setRICSPSize

◆ getRLOFOpticalFlowParameter()

virtual Ptr< RLOFOpticalFlowParameter > cv::optflow::DenseRLOFOpticalFlow::getRLOFOpticalFlowParameter ( ) const

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.getRLOFOpticalFlowParameter(		) ->	retval

Configuration of the RLOF alogrithm.

See also: optflow::RLOFOpticalFlowParameter, setRLOFOpticalFlowParameter

◆ getUsePostProc()

virtual bool cv::optflow::DenseRLOFOpticalFlow::getUsePostProc ( ) const

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.getUsePostProc(		) ->	retval

enables ximgproc::fastGlobalSmootherFilter

See also: ximgproc::fastGlobalSmootherFilter, setUsePostProc

◆ getUseVariationalRefinement()

virtual bool cv::optflow::DenseRLOFOpticalFlow::getUseVariationalRefinement ( ) const

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.getUseVariationalRefinement(		) ->	retval

enables VariationalRefinement

See also: ximgproc::fastGlobalSmootherFilter, setUsePostProc

◆ setEPICK()

virtual void cv::optflow::DenseRLOFOpticalFlow::setEPICK ( int val )

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.setEPICK(	val	) ->	None

see ximgproc::EdgeAwareInterpolator() K value.

See also: ximgproc::EdgeAwareInterpolator, getEPICK

◆ setEPICLambda()

virtual void cv::optflow::DenseRLOFOpticalFlow::setEPICLambda ( float val )

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.setEPICLambda(	val	) ->	None

see ximgproc::EdgeAwareInterpolator() lambda value.

See also: ximgproc::EdgeAwareInterpolator, getEPICLambda

◆ setEPICSigma()

virtual void cv::optflow::DenseRLOFOpticalFlow::setEPICSigma ( float val )

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.setEPICSigma(	val	) ->	None

see ximgproc::EdgeAwareInterpolator() sigma value.

See also: ximgproc::EdgeAwareInterpolator, getEPICSigma

◆ setFgsLambda()

virtual void cv::optflow::DenseRLOFOpticalFlow::setFgsLambda ( float val )

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.setFgsLambda(	val	) ->	None

see ximgproc::EdgeAwareInterpolator().

See also: ximgproc::EdgeAwareInterpolator, ximgproc::fastGlobalSmootherFilter, getFgsLambda

◆ setFgsSigma()

virtual void cv::optflow::DenseRLOFOpticalFlow::setFgsSigma ( float val )

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.setFgsSigma(	val	) ->	None

see ximgproc::EdgeAwareInterpolator().

See also: ximgproc::EdgeAwareInterpolator, ximgproc::fastGlobalSmootherFilter, getFgsSigma

◆ setForwardBackward()

virtual void cv::optflow::DenseRLOFOpticalFlow::setForwardBackward ( float val )

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.setForwardBackward(	val	) ->	None

Threshold for the forward backward confidence check.

For each grid point \( \mathbf{x} \) a motion vector \( d_{I0,I1}(\mathbf{x}) \) is computed. If the forward backward error

\[ EP_{FB} = || d_{I0,I1} + d_{I1,I0} || \]

is larger than threshold given by this function then the motion vector will not be used by the following vector field interpolation. \( d_{I1,I0} \) denotes the backward flow. Note, the forward backward test will only be applied if the threshold > 0. This may results into a doubled runtime for the motion estimation.

See also: getForwardBackward, setGridStep

◆ setGridStep()

virtual void cv::optflow::DenseRLOFOpticalFlow::setGridStep ( Size val )

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.setGridStep(	val	) ->	None

Size of the grid to spawn the motion vectors.

See also: getGridStep

◆ setInterpolation()

virtual void cv::optflow::DenseRLOFOpticalFlow::setInterpolation ( InterpolationType val )

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.setInterpolation(	val	) ->	None

Interpolation used to compute the dense optical flow.

Two interpolation algorithms are supported

INTERP_GEO applies the fast geodesic interpolation, see [103].
INTERP_EPIC_RESIDUAL applies the edge-preserving interpolation, see [225],Geistert2016.
See also
ximgproc::EdgeAwareInterpolator, getInterpolation

◆ setRICSLICType()

virtual void cv::optflow::DenseRLOFOpticalFlow::setRICSLICType ( int val )

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.setRICSLICType(	val	) ->	None

Parameter to choose superpixel algorithm variant to use:

cv::ximgproc::SLICType SLIC segments image using a desired region_size (value: 100)
cv::ximgproc::SLICType SLICO will optimize using adaptive compactness factor (value: 101)
cv::ximgproc::SLICType MSLIC will optimize using manifold methods resulting in more content-sensitive superpixels (value: 102).
See also
cv::ximgproc::createSuperpixelSLIC, cv::ximgproc::RICInterpolator

◆ setRICSPSize()

virtual void cv::optflow::DenseRLOFOpticalFlow::setRICSPSize ( int val )

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.setRICSPSize(	val	) ->	None

Parameter to tune the approximate size of the superpixel used for oversegmentation.

See also: cv::ximgproc::createSuperpixelSLIC, cv::ximgproc::RICInterpolator

◆ setRLOFOpticalFlowParameter()

virtual void cv::optflow::DenseRLOFOpticalFlow::setRLOFOpticalFlowParameter ( Ptr< RLOFOpticalFlowParameter > val )

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.setRLOFOpticalFlowParameter(	val	) ->	None

Configuration of the RLOF alogrithm.

See also: optflow::RLOFOpticalFlowParameter, getRLOFOpticalFlowParameter

◆ setUsePostProc()

virtual void cv::optflow::DenseRLOFOpticalFlow::setUsePostProc ( bool val )

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.setUsePostProc(	val	) ->	None

enables ximgproc::fastGlobalSmootherFilter

See also: getUsePostProc

◆ setUseVariationalRefinement()

virtual void cv::optflow::DenseRLOFOpticalFlow::setUseVariationalRefinement ( bool val )

pure virtual

Python:
	cv.optflow.DenseRLOFOpticalFlow.setUseVariationalRefinement(	val	) ->	None

enables VariationalRefinement

See also: getUseVariationalRefinement

The documentation for this class was generated from the following file:

opencv2/optflow/rlofflow.hpp

Public Member Functions

Static Public Member Functions

Additional Inherited Members

Detailed Description

Member Function Documentation

◆ create()

◆ getEPICK()

◆ getEPICLambda()

◆ getEPICSigma()

◆ getFgsLambda()

◆ getFgsSigma()

◆ getForwardBackward()

◆ getGridStep()

◆ getInterpolation()

◆ getRICSLICType()

◆ getRICSPSize()

◆ getRLOFOpticalFlowParameter()

◆ getUsePostProc()

◆ getUseVariationalRefinement()

◆ setEPICK()

◆ setEPICLambda()

◆ setEPICSigma()

◆ setFgsLambda()

◆ setFgsSigma()

◆ setForwardBackward()

◆ setGridStep()

◆ setInterpolation()

◆ setRICSLICType()

◆ setRICSPSize()

◆ setRLOFOpticalFlowParameter()

◆ setUsePostProc()

◆ setUseVariationalRefinement()