Package boofcv.alg.geo.f

Class EssentialResidualSampson

java.lang.Object

boofcv.alg.geo.f.EssentialResidualSampson

All Implemented Interfaces:

ModelObservationResidual<DMatrixRMaj,AssociatedPair>

public class EssentialResidualSampson
extends Object
implements ModelObservationResidual<DMatrixRMaj,AssociatedPair>

Computes the Sampson distance residual for a set of observations given an esesntial matrix. For use in least-squares non-linear optimization algorithms. Error is computed in pixels.

0 = x2^T*F*x1

E=K2'*F*K1

F=inv(K2')*E*inv(K1)

0=(K2*n2)'inv(K2)'*E*inv(K1)*(K1*n2), where n1 and n2 are normalized image coordinates

Page 287 in: R. Hartley, and A. Zisserman, "Multiple View Geometry in Computer Vision", 2nd Ed, Cambridge 2003

Constructor Summary

Constructors

Constructor	Description
EssentialResidualSampson()

Method Summary

Modifier and Type	Method	Description
double	computeResidual(AssociatedPair observation)	Compute the error for the observation
void	setCalibration1(CameraPinhole pinhole)
void	setCalibration2(CameraPinhole pinhole)
void	setModel(DMatrixRMaj E)	Specify the epipolar matrix being evaluated

Methods inherited from class java.lang.Object

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

Constructor Details

EssentialResidualSampson

public EssentialResidualSampson()

Method Details

setCalibration1

public void setCalibration1(CameraPinhole pinhole)

setCalibration2

public void setCalibration2(CameraPinhole pinhole)

setModel

public void setModel(DMatrixRMaj E)

Description copied from interface: ModelObservationResidual

Specify the epipolar matrix being evaluated

Specified by:

setModel in interface ModelObservationResidual<DMatrixRMaj,AssociatedPair>

Parameters:

E - The model being optimized

computeResidual

public double computeResidual(AssociatedPair observation)

Description copied from interface: ModelObservationResidual

Compute the error for the observation

Specified by:

computeResidual in interface ModelObservationResidual<DMatrixRMaj,AssociatedPair>

Parameters:

observation - Observation of point feature in two views

Returns:

residual Error magnitude