Class SparseOpticalFlow

Namespace

Bonsai.Vision

Assembly

Bonsai.Vision.dll

Represents an operator that calculates the optical flow for each sparse feature set in the sequence, using the iterative Lucas-Kanade method.

public class SparseOpticalFlow : Transform<Tuple<KeyPointCollection, IplImage>, KeyPointOpticalFlow>

Inheritance

object

Combinator<Tuple<KeyPointCollection, IplImage>, KeyPointOpticalFlow>

Transform<Tuple<KeyPointCollection, IplImage>, KeyPointOpticalFlow>

SparseOpticalFlow

Inherited Members

Combinator<Tuple<KeyPointCollection, IplImage>, KeyPointOpticalFlow>.Process(IObservable<Tuple<KeyPointCollection, IplImage>>)

object.ToString()

object.Equals(object)

object.Equals(object, object)

object.ReferenceEquals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

Properties

Epsilon

Gets or sets the minimum required accuracy for convergence.

public double Epsilon { get; set; }

Property Value

double

Level

Gets or sets the maximum pyramid level to use. If it is zero, pyramids are not used.

public int Level { get; set; }

Property Value

int

MaxError

Gets or sets the optional maximum allowed tracking error for each feature.

public float? MaxError { get; set; }

Property Value

float?

MaxIterations

Gets or sets the maximum number of iterations.

public int MaxIterations { get; set; }

Property Value

int

WindowSize

Gets or sets the size of the search window at each pyramid level.

public Size WindowSize { get; set; }

Property Value

Size

Methods

Process(IObservable<KeyPointCollection>)

Calculates the optical flow for each sparse feature set in an observable sequence, using the iterative Lucas-Kanade method.

public IObservable<KeyPointOpticalFlow> Process(IObservable<KeyPointCollection> source)

Parameters

source IObservable<KeyPointCollection>

A sequence of KeyPointCollection objects representing the sparse feature set over which to compute the optical flow. Each element of the sequence is compared with the previous element.

Returns

IObservable<KeyPointOpticalFlow>

A sequence of KeyPointOpticalFlow objects representing the sparse correspondences between subsequent sets of features in the original sequence.

Process(IObservable<Tuple<KeyPointCollection, IplImage>>)

Calculates the optical flow for each sparse feature set in an observable sequence, using the iterative Lucas-Kanade method, where each feature in the set is searched in the new image.

public override IObservable<KeyPointOpticalFlow> Process(IObservable<Tuple<KeyPointCollection, IplImage>> source)

Parameters

source IObservable<Tuple<KeyPointCollection, IplImage>>

A sequence of pairs where the first item is a KeyPointCollection object representing the set of features to find, and the second item is a target image on which the algorithm will try to find the features.

Returns

IObservable<KeyPointOpticalFlow>

A sequence of KeyPointOpticalFlow objects representing the sparse correspondences between each set of features in the sequence and a target image.