#include <ProbabilisticPrimarySceneObjectVisualization.h>

Public Member Functions
void	addTermScore (double pScore)

void	appendVisualizer (boost::shared_ptr< ProbabilisticSecondarySceneObjectVisualization > pVisualizer)

void	drawInInferenceMode (const std::string pScene, const std_msgs::ColorRGBA &pColor, unsigned int &pMarkerId)

void	drawInLearningMode (const std::string pScene, const std_msgs::ColorRGBA &pColor)

void	drawInLearningTrajectoryMode (const std::string pScene)

void	drawInTargetingMode (const std::string pScene, const std_msgs::ColorRGBA &pColor)

std::string	getSceneObjectName ()

void	normalizeHypothesisScore (unsigned int pNumberOfSlots)

	ProbabilisticPrimarySceneObjectVisualization (std::string pName)

void	resetHypothesis ()

void	resetTermScores ()

void	setBestPoseCandidate (boost::shared_ptr< ISM::Pose > pPose)

void	setBestStatus (bool pStatus)

void	setDrawingParameters (const double pScale, const float pSigmaMultiplicator, const std::string pFrameId)

void	setHypothesisScore (double pHypothesisScore)

void	setLearningSamples (const std::vector< Eigen::Vector3d > &pSamples)

void	setPose (boost::shared_ptr< ISM::Pose > pPose)

	~ProbabilisticPrimarySceneObjectVisualization ()

Private Attributes
boost::shared_ptr< ISM::Pose >	mAbsolutePose

double	mAppearanceScore

boost::shared_ptr< ISM::Pose >	mBestCandidatePose

double	mBestHypothesisScore

boost::shared_ptr< ISM::Pose >	mBestPose

std::vector< double >	mBestTermScores

double	mExistenceScore

ros::NodeHandle	mHandle

boost::shared_ptr< ros::Publisher >	mPublisher

std::vector< Eigen::Vector3d >	mSamples

std::string	mSceneObject

std::vector< boost::shared_ptr< ProbabilisticSecondarySceneObjectVisualization > >	mSceneObjectVisualizers

double	mShapeScore

std::vector< double >	mTermScores

CoordinateFrameVisualizer	visualizerFrame

KinematicChainVisualizer	visualizerKinematics

SampleVisualizer	visualizerSamples

Detailed Description

Visualizer class for primary scenes objects.

Author: Joachim Gehrung

Version: See SVN

Definition at line 58 of file ProbabilisticPrimarySceneObjectVisualization.h.

Constructor & Destructor Documentation

Visualization::ProbabilisticPrimarySceneObjectVisualization::ProbabilisticPrimarySceneObjectVisualization ( std::string pName )

Constructor.

Parameters

pName The name of this primary scene object.

Definition at line 22 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

Visualization::ProbabilisticPrimarySceneObjectVisualization::~ProbabilisticPrimarySceneObjectVisualization ( )

Destructor.

Definition at line 30 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

Member Function Documentation

void Visualization::ProbabilisticPrimarySceneObjectVisualization::addTermScore ( double pScore )

Adds the score of one of the terms that form the best hypothesis

Parameters

pScore	The score of the appearance term.
pShapeScore	The score of the shape term.
pExistenceScore	The score of the existence term.

Definition at line 243 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

void Visualization::ProbabilisticPrimarySceneObjectVisualization::appendVisualizer ( boost::shared_ptr< ProbabilisticSecondarySceneObjectVisualization > pVisualizer )

Appends a scene object visualizer.

Parameters

pVisualizer The scene object visualizer to append.

Definition at line 34 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

void Visualization::ProbabilisticPrimarySceneObjectVisualization::drawInInferenceMode	(	const std::string	pScene,
		const std_msgs::ColorRGBA &	pColor,
		unsigned int &	pMarkerId
	)

Draws the kinematic chain for the best scoring hypothesis regarding the assignment of evidence to the slots. A chain for every primary scene object for every scene will be visualized in a separate topic. One ring color implies the primary scene object, the other the secondary scene object.

Parameters

pScene	The name of the scene. Required for creating un unique frame id.
pColor	The color indicating the scene object.
pMarkerId	Serves as unique id for the marker. Any marker sent with the same id will overwrite the old one.

Definition at line 119 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

void Visualization::ProbabilisticPrimarySceneObjectVisualization::drawInLearningMode	(	const std::string	pScene,
		const std_msgs::ColorRGBA &	pColor
	)

Draws every primary scene object in a distinct coordinate frame. One ring color implies the primary scene object, the other the secondary scene object.

Parameters

pScene	The name of the scene. Required for creating un unique frame id.
pColor	The color indicating the scene object.

Definition at line 165 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

void Visualization::ProbabilisticPrimarySceneObjectVisualization::drawInLearningTrajectoryMode ( const std::string pScene )

Draws the object trajectories used for learning.

Parameters

pScene The name of the scene. Required for creating un unique frame id.

Definition at line 202 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

void Visualization::ProbabilisticPrimarySceneObjectVisualization::drawInTargetingMode	(	const std::string	pScene,
		const std_msgs::ColorRGBA &	pColor
	)

Draws the probability distribuion indicating the position of the other scene objects. One ring color implies the primary scene object, the other the secondary scene object.

Parameters

pScene	The name of the scene. Required for creating un unique frame id.
pColor	The color indicating the scene object.

Definition at line 78 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

std::string Visualization::ProbabilisticPrimarySceneObjectVisualization::getSceneObjectName ( )

Returns the name of the primary scene object associated with this visualizer.

Returns: The name of the primary scene object.

Definition at line 238 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

void Visualization::ProbabilisticPrimarySceneObjectVisualization::normalizeHypothesisScore ( unsigned int pNumberOfSlots )

Normalizes the score for the best hypothesis.

Parameters

pNumberOfSlots The number of slots used for normalization.

Definition at line 270 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

void Visualization::ProbabilisticPrimarySceneObjectVisualization::resetHypothesis ( )

Resets the candidates and positions.

Definition at line 275 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

void Visualization::ProbabilisticPrimarySceneObjectVisualization::resetTermScores ( )

Resets all term scores.

Definition at line 248 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

void Visualization::ProbabilisticPrimarySceneObjectVisualization::setBestPoseCandidate ( boost::shared_ptr< ISM::Pose > pPose )

The best absolute pose of the primary object. It was determined by the best scoring hypothesis regarding the assignment of evidence to the slots.

Parameters

pPose Best absolute pose of the primary scene object.

Definition at line 63 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

void Visualization::ProbabilisticPrimarySceneObjectVisualization::setBestStatus ( bool pStatus )

Marks the scene object with the best score.

Parameters

pStatus True, to select the scene object as the one with the best score.

Definition at line 281 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

void Visualization::ProbabilisticPrimarySceneObjectVisualization::setDrawingParameters	(	const double	pScale,
		const float	pSigmaMultiplicator,
		const std::string	pFrameId
	)

Sets the parameter requried for visualizing the secondary scene object.

Parameters

pScale	Factor to multiply the kernel with.
pSigmaMultiplicator	Scaling factor for the size of the visualized covariance ellipsoid.
pFrameId	The name of the coordinate frame that should be drawn into.

Definition at line 39 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

void Visualization::ProbabilisticPrimarySceneObjectVisualization::setHypothesisScore ( double pHypothesisScore )

If the given score is better than the last best score, the last best pose is overwritten by the last pose.

Parameters

pHypothesisScore The score of the last hypothesis.

Definition at line 253 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

void Visualization::ProbabilisticPrimarySceneObjectVisualization::setLearningSamples ( const std::vector< Eigen::Vector3d > & pSamples )

Sets the list of samples used for learning the gaussian mixture distribution.

Parameters

pSamples The list of samples used for learning the gaussian mixture distribution.

Definition at line 73 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

void Visualization::ProbabilisticPrimarySceneObjectVisualization::setPose ( boost::shared_ptr< ISM::Pose > pPose )

Sets the absolute pose of the primary object associated with this visualizer.

Parameters

pPose Absolute pose of the primary scene object.

Definition at line 53 of file ProbabilisticPrimarySceneObjectVisualization.cpp.

Member Data Documentation

boost::shared_ptr<ISM::Pose> Visualization::ProbabilisticPrimarySceneObjectVisualization::mAbsolutePose

private

The absolute pose of the primary object associated with this visualizer.

Definition at line 225 of file ProbabilisticPrimarySceneObjectVisualization.h.

double Visualization::ProbabilisticPrimarySceneObjectVisualization::mAppearanceScore

private

The score of the single terms the best hypothesis score is made of.

Definition at line 240 of file ProbabilisticPrimarySceneObjectVisualization.h.

boost::shared_ptr<ISM::Pose> Visualization::ProbabilisticPrimarySceneObjectVisualization::mBestCandidatePose

private

Candidate for the best absolute position of the primary object. It was determined by the best scoring hypothesis regarding the assignment of evidence to the slots.

Definition at line 250 of file ProbabilisticPrimarySceneObjectVisualization.h.

double Visualization::ProbabilisticPrimarySceneObjectVisualization::mBestHypothesisScore

private

The score of the best hypothesis.

Definition at line 230 of file ProbabilisticPrimarySceneObjectVisualization.h.

boost::shared_ptr<ISM::Pose> Visualization::ProbabilisticPrimarySceneObjectVisualization::mBestPose

private

Best absolute pose of the primary object. It was determined by the best scoring hypothesis regarding the assignment of evidence to the slots.

Definition at line 245 of file ProbabilisticPrimarySceneObjectVisualization.h.

std::vector<double> Visualization::ProbabilisticPrimarySceneObjectVisualization::mBestTermScores

private

Definition at line 235 of file ProbabilisticPrimarySceneObjectVisualization.h.

double Visualization::ProbabilisticPrimarySceneObjectVisualization::mExistenceScore

private

Definition at line 240 of file ProbabilisticPrimarySceneObjectVisualization.h.

ros::NodeHandle Visualization::ProbabilisticPrimarySceneObjectVisualization::mHandle

private

Node handle for inintializing the publisher.

Definition at line 200 of file ProbabilisticPrimarySceneObjectVisualization.h.

boost::shared_ptr<ros::Publisher> Visualization::ProbabilisticPrimarySceneObjectVisualization::mPublisher

private

Publisher for all visualization messages generated in this class.

Definition at line 205 of file ProbabilisticPrimarySceneObjectVisualization.h.

std::vector<Eigen::Vector3d> Visualization::ProbabilisticPrimarySceneObjectVisualization::mSamples

private

The list of samples used for learning the gaussian mixture distribution.

Definition at line 260 of file ProbabilisticPrimarySceneObjectVisualization.h.

std::string Visualization::ProbabilisticPrimarySceneObjectVisualization::mSceneObject

private

The name of this primary scene object.

Definition at line 195 of file ProbabilisticPrimarySceneObjectVisualization.h.

std::vector<boost::shared_ptr<ProbabilisticSecondarySceneObjectVisualization> > Visualization::ProbabilisticPrimarySceneObjectVisualization::mSceneObjectVisualizers

private

A list of visualizers for all appended secondary scene objects.

Definition at line 255 of file ProbabilisticPrimarySceneObjectVisualization.h.

double Visualization::ProbabilisticPrimarySceneObjectVisualization::mShapeScore

private

Definition at line 240 of file ProbabilisticPrimarySceneObjectVisualization.h.

std::vector<double> Visualization::ProbabilisticPrimarySceneObjectVisualization::mTermScores

private

A list of the current term scores and the best term scores.

Definition at line 235 of file ProbabilisticPrimarySceneObjectVisualization.h.

CoordinateFrameVisualizer Visualization::ProbabilisticPrimarySceneObjectVisualization::visualizerFrame

private