Abstract base class for message passing inference algorithmes. More...

#include <MessagePassing.h>

Inheritance diagram for DirectGraphicalModels::CMessagePassing:

[legend]

Collaboration diagram for DirectGraphicalModels::CMessagePassing:

Public Member Functions
	CMessagePassing (CGraphPairwise &graph)
	Constructor. More...

virtual	~CMessagePassing (void)

virtual void	infer (unsigned int nIt=1)
	Inference. More...

Public Member Functions inherited from DirectGraphicalModels::CInfer
	CInfer (CGraph &graph)
	Constructor. More...

	CInfer (const CInfer &)=delete

virtual	~CInfer ()=default

const CInfer &	operator= (const CInfer &)=delete

vec_byte_t	decode (unsigned int nIt=0, Mat &lossMatrix=EmptyMat)
	Approximate decoding. More...

vec_float_t	getConfidence (void) const
	Returns the confidence of the perdiction. More...

vec_float_t	getPotentials (byte state) const
	Returns the potnetials for the selected state (class) More...

Protected Member Functions
CGraphPairwise &	getGraphPairwise (void) const
	Returns the graph. More...

virtual void	calculateMessages (unsigned int nIt)=0
	Calculates messages, associated with the edges of corresponding graphical model. More...

void	calculateMessage (Edge edge, float temp, float *&dst, bool maxSum=false)
	Calculates one message for the specified edge edge. More...

void	createMessages (void)
	Allocates memory for Edge::msg and Edge::msg_temp containers for all edges in the graph. More...

void	deleteMessages (void)
	Deletes memory for Edge::msg and Edge::msg_temp containers for all edges in the graph. More...

void	swapMessages (void)
	Swaps Edge::msg and Edge::msg_temp for all edges in the graph. More...

Protected Member Functions inherited from DirectGraphicalModels::CInfer
CGraph &	getGraph (void) const
	Returns the reference to the graph. More...

Static Protected Member Functions
static float	MatMul (const Mat &M, const float v, float &dst, bool maxSum=false)
	Specific matrix multiplication. More...

Detailed Description

Abstract base class for message passing inference algorithmes.

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 18 of file MessagePassing.h.

Constructor & Destructor Documentation

◆ CMessagePassing()

DirectGraphicalModels::CMessagePassing::CMessagePassing ( CGraphPairwise & graph )

inline

Constructor.

Parameters

graph The graph

Definition at line 25 of file MessagePassing.h.

◆ ~CMessagePassing()

virtual DirectGraphicalModels::CMessagePassing::~CMessagePassing ( void )

inlinevirtual

Definition at line 26 of file MessagePassing.h.

Member Function Documentation

◆ calculateMessage()

void DirectGraphicalModels::CMessagePassing::calculateMessage	(	Edge *	edge,
		float *	temp,
		float *&	dst,
		bool	maxSum = `false`
	)

protected

Calculates one message for the specified edge edge.

PPL-safe function.

Parameters

[in]	edge	Graph edge
[in]	temp	Auxilary array of nStates values. Introduced for higher perfomance reasons.
[out]	dst	Destination array for calculated message. Usually edge->msg or edge->msg_temp. If the pointer is NULL, destination container will be created as a new vector of length nStates.
[in]	maxSum	Flag indicating weather the message must be calculated according to the sum-product (false) or max-product (true) algorithm.

Definition at line 82 of file MessagePassing.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ calculateMessages()

virtual void DirectGraphicalModels::CMessagePassing::calculateMessages ( unsigned int nIt )

protectedpure virtual

Calculates messages, associated with the edges of corresponding graphical model.

This function may modify Edge::msg and Edge::msg_temp containers of graph edges

Parameters

nIt	Number of iterations

Implemented in DirectGraphicalModels::CInferChain, DirectGraphicalModels::CInferTree, DirectGraphicalModels::CInferTRW, and DirectGraphicalModels::CInferLBP.

Here is the caller graph for this function:

◆ createMessages()

void DirectGraphicalModels::CMessagePassing::createMessages ( void )

protected

Allocates memory for Edge::msg and Edge::msg_temp containers for all edges in the graph.

Definition at line 113 of file MessagePassing.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ deleteMessages()

void DirectGraphicalModels::CMessagePassing::deleteMessages ( void )

protected

Deletes memory for Edge::msg and Edge::msg_temp containers for all edges in the graph.

Definition at line 129 of file MessagePassing.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ getGraphPairwise()

CGraphPairwise& DirectGraphicalModels::CMessagePassing::getGraphPairwise ( void ) const

inlineprotected

Returns the graph.

Returns: The graph

Definition at line 36 of file MessagePassing.h.

Here is the call graph for this function:

Here is the caller graph for this function:

◆ infer()

void DirectGraphicalModels::CMessagePassing::infer ( unsigned int nIt = 1 )

virtual

Inference.

This function estimates the marginal potentials for each graph node, and stores them as node potentials

This function modifies Node::Pot containers of graph nodes

Parameters

nIt	Number of iterations

Note: This function must not to be linear, i.e. \( infer(\alpha\times N)\not\equiv\alpha\times infer(N) \); This function substitutes the graph nodes' potentials with estimated marginal potentials

Implements DirectGraphicalModels::CInfer.

Reimplemented in DirectGraphicalModels::CInferTRW.

Definition at line 7 of file MessagePassing.cpp.

Here is the call graph for this function:

◆ MatMul()

float DirectGraphicalModels::CMessagePassing::MatMul	(	const Mat &	M,
		const float *	v,
		float *&	dst,
		bool	maxSum = `false`
	)

staticprotected

Specific matrix multiplication.

This function calculates the result of multiplying square of matrix M by vector v as following: \(\vec{dst} = (M\cdot M)^\top\times\vec{v}\)

Parameters

[in]	M	Matrix: Mat(size: M.height x M.width; type: CV_32FC1)
[in]	v	Vector of length M.height
[out]	dst	Resulting vector. If the pointer is NULL, destination container will be created as a new vector of length M.width.
[in]	maxSum	Flag indicating weather the max-sum multiplication should be performed

Returns: The sum of all elemts in vector dst

Definition at line 157 of file MessagePassing.cpp.

Here is the caller graph for this function:

◆ swapMessages()

void DirectGraphicalModels::CMessagePassing::swapMessages ( void )

protected

Swaps Edge::msg and Edge::msg_temp for all edges in the graph.

Definition at line 147 of file MessagePassing.cpp.

Here is the call graph for this function:

Here is the caller graph for this function:

The documentation for this class was generated from the following files:

D:/Projects/DGM/modules/DGM/MessagePassing.h
D:/Projects/DGM/modules/DGM/MessagePassing.cpp

Public Member Functions

Protected Member Functions

Static Protected Member Functions

Detailed Description

Constructor & Destructor Documentation

◆ CMessagePassing()

◆ ~CMessagePassing()

Member Function Documentation

◆ calculateMessage()

◆ calculateMessages()

◆ createMessages()

◆ deleteMessages()

◆ getGraphPairwise()

◆ infer()

◆ MatMul()

◆ swapMessages()