Potts edge training class. More...

#include <TrainEdgePotts.h>

Inheritance diagram for DirectGraphicalModels::CTrainEdgePotts:

[legend]

Collaboration diagram for DirectGraphicalModels::CTrainEdgePotts:

Public Member Functions
	CTrainEdgePotts (byte nStates, word nFeatures)
	Constructor. More...

virtual	~CTrainEdgePotts (void)

virtual void	reset (void)
	Resets class variables. More...

virtual void	addFeatureVecs (const Mat &featureVector1, byte gt1, const Mat &featureVector2, byte gt2)
	Adds a pair of feature vectors. More...

Public Member Functions inherited from DirectGraphicalModels::CTrainEdge
	CTrainEdge (byte nStates, word nFeatures)
	Constructor. More...

virtual	~CTrainEdge (void)

void	addFeatureVecs (const Mat &featureVectors, const Mat &gt, const CGraphExt *pGraph)
	Adds a block of new feature vectors. More...

void	addFeatureVecs (const vec_mat_t &featureVectors, const Mat &gt, const CGraphExt *pGraph)
	Adds a block of new feature vectors. More...

virtual void	train (bool doClean=false)
	Random model training. More...

Mat	getEdgePotentials (const Mat &featureVector1, const Mat &featureVector2, float *params, size_t params_len, float weight=1.0f) const
	Returns the edge potential, based on the feature vectors. More...

Public Member Functions inherited from DirectGraphicalModels::ITrain
	ITrain (byte nStates, word nFeatures)
	Constructor. More...

virtual	~ITrain (void)

word	getNumFeatures (void) const
	Returns number of features. More...

Public Member Functions inherited from DirectGraphicalModels::CBaseRandomModel
	CBaseRandomModel (byte nStates)
	Constructor. More...

virtual	~CBaseRandomModel (void)

virtual void	save (const std::string &path, const std::string &name=std::string(), short idx=-1) const
	Saves the training data. More...

virtual void	load (const std::string &path, const std::string &name=std::string(), short idx=-1)
	Loads the training data. More...

byte	getNumStates (void) const
	Returns number of features. More...

Static Public Member Functions
static Mat	getEdgePotentials (float val, byte nStates)
	Returns the data-independent edge potentials. More...

static Mat	getEdgePotentials (float *values, byte nStates)
	Returns the data-independent edge potentials. More...

Protected Member Functions
virtual void	saveFile (FILE *pFile) const
	Saves the random model into the file. More...

virtual void	loadFile (FILE *pFile)
	Loads the random model from the file. More...

virtual Mat	calculateEdgePotentials (const Mat &featureVector1, const Mat &featureVector2, float *params, size_t params_len) const
	Returns the data-independent edge potentials. More...

Protected Member Functions inherited from DirectGraphicalModels::CBaseRandomModel
std::string	generateFileName (const std::string &path, const std::string &name, short idx) const
	Generates name of the data file for storing random model parameters. More...

Additional Inherited Members
Protected Attributes inherited from DirectGraphicalModels::ITrain
word	m_nFeatures
	The number of features (length of the feature vector) More...

Protected Attributes inherited from DirectGraphicalModels::CBaseRandomModel
byte	m_nStates
	The number of states (classes) More...

Detailed Description

Potts edge training class.

This class implements the Potts model for edge (pairwise) potentials, which is both training- and test-data-independent, and thus may be applied without edge training procedure, i.e. functions reset(), save(), load(), addFeatureVecs() are unnecessary here.

This class may be used in case when the training- and test-data for edges is not available.

Author: Sergey G. Kosov, serge.nosp@m.y.ko.nosp@m.sov@p.nosp@m.roje.nosp@m.ct-10.nosp@m..de

Definition at line 19 of file TrainEdgePotts.h.

Constructor & Destructor Documentation

DirectGraphicalModels::CTrainEdgePotts::CTrainEdgePotts	(	byte	nStates,
		word	nFeatures
	)

inline

Constructor.

Parameters

nStates	Number of states (classes)
nFeatures	Number of features

Definition at line 27 of file TrainEdgePotts.h.

virtual DirectGraphicalModels::CTrainEdgePotts::~CTrainEdgePotts ( void )

inlinevirtual

Definition at line 28 of file TrainEdgePotts.h.

Member Function Documentation

virtual void DirectGraphicalModels::CTrainEdgePotts::addFeatureVecs	(	const Mat &	featureVector1,
		byte	gt1,
		const Mat &	featureVector2,
		byte	gt2
	)

inlinevirtual

Adds a pair of feature vectors.

Used to add featureVector1 and featureVector2, corresponding to the ground-truth states (classes) gt1 and gt2 for training. Here the couple {featureVector1, gt1} corresponds to the first node of the edge, and the couple {featureVector2, gt2} - to the second node.

Parameters

featureVector1	Multi-dimensinal point: Mat(size: nFeatures x 1; type: CV_8UC1), corresponding to the first node of the edge.
gt1	The ground-truth state (class) of the first node of the edge, given by featureVector1
featureVector2	Multi-dimensinal point: Mat(size: nFeatures x 1; type: CV_8UC1), corresponding to the second node of the edge.
gt2	The ground-truth state (class) of the second node of the edge, given by featureVector2

Implements DirectGraphicalModels::CTrainEdge.

Reimplemented in DirectGraphicalModels::CTrainEdgePrior.

Definition at line 32 of file TrainEdgePotts.h.

Mat DirectGraphicalModels::CTrainEdgePotts::calculateEdgePotentials	(	const Mat &	featureVector1,
		const Mat &	featureVector2,
		float *	params,
		size_t	params_len
	)		const

protectedvirtual

Returns the data-independent edge potentials.

This function returns matrix with diagonal elements equal to parameter \(\vec{\theta}\) provided through argument params; all the other elements are 1's - this imitates the Potts model.

\[edgePot[nStates][nStates] = \begin{bmatrix} \theta_1 & 1 & \cdots & 1 \\ 1 & \theta_2 & \cdots & 1 \\ \vdots & \vdots & \ddots & \vdots \\ 1 & 1 & \cdots & \theta_{nStates} \end{bmatrix} \]

Parameters

featureVector1	Multi-dimensinal point \(\textbf{f}_1\): Mat(size: nFeatures x 1; type: CV_{XX}C1), corresponding to the first node of the edge It is not used in the Potts model, thus may be empty Mat()
featureVector2	Multi-dimensinal point \(\textbf{f}_2\): Mat(size: nFeatures x 1; type: CV_{XX}C1), corresponding to the second node of the edge It is not used in the Potts model, thus may be empty Mat()
params	Array of control parameters \(\vec{\theta}\), which may consist either from one parameter (in this case all the diagonal elemets will be the same), or from nStates parameters, specifying smoothness strength for each state (class) individually.
params_len	The length of the params parameter. It must be equal to (1 \|\| nStates).

Returns: The edge potential matrix: Mat(size: nStates x nStates; type: CV_32FC1)

Implements DirectGraphicalModels::CTrainEdge.

Reimplemented in DirectGraphicalModels::CTrainEdgePottsCS, and DirectGraphicalModels::CTrainEdgePrior.

Definition at line 23 of file TrainEdgePotts.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

Mat DirectGraphicalModels::CTrainEdgePotts::getEdgePotentials	(	float	val,
		byte	nStates
	)

static

Returns the data-independent edge potentials.

This function returns matrix with diagonal elements equal to the argument val, all the other elements are 1's, what imitates the Potts model.

\[edgePot[nStates][nStates] = \begin{bmatrix} val & 1 & \cdots & 1 \\ 1 & val & \cdots & 1 \\ \vdots & \vdots & \ddots & \vdots \\ 1 & 1 & \cdots & val \end{bmatrix} \]

Parameters

val	The diagonal element of the matrix
nStates	Number of states (classes)

Returns: The edge potential matrix: Mat(size: nStates x nStates; type: CV_32FC1)

Definition at line 8 of file TrainEdgePotts.cpp.

Here is the caller graph for this function:

Mat DirectGraphicalModels::CTrainEdgePotts::getEdgePotentials	(	float *	values,
		byte	nStates
	)

static

Returns the data-independent edge potentials.

This function returns matrix with diagonal elements specified by argument array values, all the other elements are 1's, what imitates the Potts model.

\[edgePot[nStates][nStates] = \begin{bmatrix} values_1 & 1 & \cdots & 1 \\ 1 & values_2 & \cdots & 1 \\ \vdots & \vdots & \ddots & \vdots \\ 1 & 1 & \cdots & values_{nStates} \end{bmatrix} \]

Parameters

values	The array of diagonal elements of the matrix. In must include nStates values, specifying smoothness strength for each state (class) individually.
nStates	Number of states (classes)