Class RICInterpolator

Namespace

OpenCvSharp.XImgProc

Assembly

OpenCvSharp.dll

Sparse match interpolation algorithm based on modified piecewise locally-weighted affine estimator called Robust Interpolation method of Correspondences (RIC), and Variational and Fast Global Smoother as post-processing filter. RICInterpolator is an extension of EdgeAwareInterpolator; its main concept is a piece-wise affine model based on over-segmentation via SLIC superpixel estimation, with an efficient propagation mechanism to estimate among the piece-wise models.

public class RICInterpolator : SparseMatchInterpolator, IDisposable

Inheritance

object

CvObject

CvPtrObject

Algorithm

SparseMatchInterpolator

RICInterpolator

Implements

IDisposable

Inherited Members

SparseMatchInterpolator.Interpolate(InputArray, InputArray, InputArray, InputArray, OutputArray)

Algorithm.Write(FileStorage)

Algorithm.Read(FileNode)

Algorithm.Empty

Algorithm.Save(string)

Algorithm.GetDefaultName()

CvPtrObject.RawPtr

CvObject.IsDisposed

CvObject.ptr

CvObject.SetSafeHandle(OpenCvSafeHandle)

CvObject.Dispose()

CvObject.Dispose(bool)

CvObject.DisposeManaged()

CvObject.DisposeUnmanaged()

CvObject.ThrowIfDisposed()

CvObject.CvPtr

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Properties

Alpha

A global weight for transforming geodesic distance into weight.

public float Alpha { get; set; }

Property Value

float

FGSLambda

The respective fastGlobalSmootherFilter() lambda parameter.

public float FGSLambda { get; set; }

Property Value

float

FGSSigma

The respective fastGlobalSmootherFilter() sigma parameter.

public float FGSSigma { get; set; }

Property Value

float

K

K is a number of nearest-neighbor matches considered, when fitting a locally affine model for a superpixel segment. Lower values would make the interpolation noticeably faster. The original implementation uses 32.

public int K { get; set; }

Property Value

int

MaxFlow

A threshold to validate the predictions using a certain piece-wise affine model. If the prediction exceeds the threshold the translational model will be applied instead.

public float MaxFlow { get; set; }

Property Value

float

ModelIter

The number of iterations for piece-wise affine model estimation.

public int ModelIter { get; set; }

Property Value

int

RefineModels

Whether additional refinement of the piece-wise affine models is employed.

public bool RefineModels { get; set; }

Property Value

bool

SuperpixelMode

Chooses the superpixel algorithm variant to use: SLIC (100) segments the image using a desired region size, SLICO (101) optimizes using an adaptive compactness factor, and MSLIC (102) optimizes using manifold methods resulting in more content-sensitive superpixels.

public int SuperpixelMode { get; set; }

Property Value

int

SuperpixelNNCnt

The number of nearest-neighbor matches for each superpixel considered, when fitting a locally affine model.

public int SuperpixelNNCnt { get; set; }

Property Value

int

SuperpixelRuler

Tunes the enforcement of the superpixel smoothness factor used for oversegmentation.

public float SuperpixelRuler { get; set; }

Property Value

float

SuperpixelSize

The approximate size of the superpixel used for oversegmentation.

public int SuperpixelSize { get; set; }

Property Value

int

UseGlobalSmootherFilter

Whether the fastGlobalSmootherFilter() post-processing is employed.

public bool UseGlobalSmootherFilter { get; set; }

Property Value

bool

UseVariationalRefinement

Whether the VariationalRefinement post-processing is employed.

public bool UseVariationalRefinement { get; set; }

Property Value

bool

Methods

Create()

Factory method that creates an instance of the RICInterpolator.

public static RICInterpolator Create()

Returns

RICInterpolator

SetCostMap(Mat)

Interface to provide a more elaborated cost map, i.e. edge map, for the edge-aware term. This implementation is based on a rather simple gradient-based edge map estimation. To use a more complex edge map estimator (e.g. StructuredEdgeDetection) that may lead to improved accuracies, the internal edge map estimation can be bypassed here.

public void SetCostMap(Mat costMap)

Parameters

costMap Mat

a type CV_32FC1 Mat is required.