Sparse stereo matching using block matching. More...

#include <Stereo.h>

Inheritance diagram for rtabmap::Stereo:

Public Member Functions
	Stereo (const ParametersMap &parameters=ParametersMap())
	Constructor.

virtual	~Stereo ()
	Virtual destructor.

virtual void	parseParameters (const ParametersMap &parameters)
	Parse parameters from a parameters map.

virtual std::vector< cv::Point2f >	computeCorrespondences (const cv::Mat &leftImage, const cv::Mat &rightImage, const std::vector< cv::Point2f > &leftCorners, std::vector< unsigned char > &status) const
	Compute stereo correspondences using block matching.

cv::Size	winSize () const
	Get the search window size.

int	iterations () const
	Get the number of iterations.

int	maxLevel () const
	Get the maximum pyramid level.

float	minDisparity () const
	Get the minimum disparity value.

float	maxDisparity () const
	Get the maximum disparity value.

bool	winSSD () const
	Check if SSD (Sum of Squared Differences) is used.

virtual bool	isGpuEnabled () const
	Check if GPU acceleration is enabled.

Static Public Member Functions
static Stereo *	create (const ParametersMap &parameters=ParametersMap())
	Factory method to create a Stereo instance.

Detailed Description

Sparse stereo matching using block matching.

This class implements sparse stereo matching to find corresponding feature points between stereo image pairs using block matching with a search window. Unlike dense stereo matching, this class works with sparse feature points rather than computing disparity for every pixel.

The algorithm uses a pyramidal approach for efficiency, searching for correspondences within a specified disparity range using either SAD (Sum of Absolute Differences) or SSD (Sum of Squared Differences) as the matching cost.

Note: Input images must be grayscale (CV_8UC1).

See also: StereoOpticalFlow for an alternative implementation using optical flow; StereoDense for dense stereo matching

Definition at line 55 of file Stereo.h.

Constructor & Destructor Documentation

◆ Stereo()

rtabmap::Stereo::Stereo ( const ParametersMap & parameters = ParametersMap() )

Constructor.

Initializes a Stereo instance with default parameter values and then parses the provided parameters map to override defaults.

Parameters

parameters Optional parameters map containing configuration values. If empty, default values are used.

◆ ~Stereo()

virtual rtabmap::Stereo::~Stereo ( )

inlinevirtual

Virtual destructor.

Definition at line 87 of file Stereo.h.

Member Function Documentation

◆ create()

static Stereo * rtabmap::Stereo::create ( const ParametersMap & parameters = ParametersMap() )

static

Factory method to create a Stereo instance.

Creates a Stereo instance based on the stereo optical flow parameter in the provided parameters map. If optical flow is enabled, creates a StereoOpticalFlow instance; otherwise, creates a standard Stereo instance.

Parameters

parameters Parameters map containing configuration values. The Parameters::kStereoOpticalFlow() parameter determines which implementation to create.

Returns: Pointer to the created Stereo instance (caller owns the memory). Returns StereoOpticalFlow if optical flow is enabled, otherwise Stereo.

◆ parseParameters()

virtual void rtabmap::Stereo::parseParameters ( const ParametersMap & parameters )

virtual

Parse parameters from a parameters map.

Updates the algorithm's configuration based on the provided parameters map. Supported parameters:

Parameters::kStereoWinWidth() - Search window width
Parameters::kStereoWinHeight() - Search window height
Parameters::kStereoIterations() - Number of iterations
Parameters::kStereoMaxLevel() - Maximum pyramid level
Parameters::kStereoMinDisparity() - Minimum disparity value
Parameters::kStereoMaxDisparity() - Maximum disparity value
Parameters::kStereoSSD() - Use SSD instead of SAD

Parameters

parameters Parameters map containing configuration values

Reimplemented in rtabmap::StereoOpticalFlow.

◆ computeCorrespondences()

virtual std::vector< cv::Point2f > rtabmap::Stereo::computeCorrespondences	(	const cv::Mat &	leftImage,
		const cv::Mat &	rightImage,
		const std::vector< cv::Point2f > &	leftCorners,
		std::vector< unsigned char > &	status
	)		const

virtual

Compute stereo correspondences using block matching.

Finds corresponding points in the right stereo image for the given points in the left stereo image using block matching with a search window. The algorithm uses a pyramidal approach for efficiency.

Parameters

leftImage	Left stereo image (must be CV_8UC1 grayscale)
rightImage	Right stereo image (must be CV_8UC1 grayscale)
leftCorners	Input vector of feature points in the left image
status	Output vector indicating which correspondences are valid (1) or invalid (0). The size matches leftCorners.size().

Returns: Vector of corresponding points in the right image. The size matches leftCorners.size(). Invalid correspondences may have coordinates outside the image bounds.

Note: Both input images must be grayscale (CV_8UC1).; The algorithm searches for correspondences within the disparity range [minDisparity(), maxDisparity()] and uses a search window of size winSize().

Reimplemented in rtabmap::StereoOpticalFlow.

◆ winSize()

cv::Size rtabmap::Stereo::winSize ( ) const

inline

Get the search window size.

Returns: Search window size as cv::Size(winWidth_, winHeight_)

Definition at line 154 of file Stereo.h.

◆ iterations()

int rtabmap::Stereo::iterations ( ) const

inline

Get the number of iterations.

Returns: Number of iterations for the matching algorithm

Definition at line 160 of file Stereo.h.

◆ maxLevel()

int rtabmap::Stereo::maxLevel ( ) const

inline

Get the maximum pyramid level.

Returns: Maximum pyramid level used in the pyramidal approach

Definition at line 166 of file Stereo.h.

◆ minDisparity()

float rtabmap::Stereo::minDisparity ( ) const

inline

Get the minimum disparity value.

Returns: Minimum disparity value to search (in pixels)

Definition at line 172 of file Stereo.h.

◆ maxDisparity()

float rtabmap::Stereo::maxDisparity ( ) const

inline

Get the maximum disparity value.

Returns: Maximum disparity value to search (in pixels)

Definition at line 178 of file Stereo.h.

◆ winSSD()

bool rtabmap::Stereo::winSSD ( ) const

inline

Check if SSD (Sum of Squared Differences) is used.

Returns: true if SSD is used, false if SAD (Sum of Absolute Differences) is used

Definition at line 184 of file Stereo.h.

◆ isGpuEnabled()

virtual bool rtabmap::Stereo::isGpuEnabled ( ) const

inlinevirtual

Check if GPU acceleration is enabled.

Returns: Always returns false for the base Stereo class

Reimplemented in rtabmap::StereoOpticalFlow.

Definition at line 190 of file Stereo.h.

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

corelib/include/rtabmap/core/Stereo.h

Public Member Functions

Static Public Member Functions

Detailed Description

Constructor & Destructor Documentation

◆ Stereo()

◆ ~Stereo()

Member Function Documentation

◆ create()

◆ parseParameters()

◆ computeCorrespondences()

◆ winSize()

◆ iterations()

◆ maxLevel()

◆ minDisparity()

◆ maxDisparity()

◆ winSSD()

◆ isGpuEnabled()