cv::detail::BundleAdjusterBase Class Reference

Base class for all camera parameters refinement methods. More...

#include <opencv2/stitching/detail/motion_estimators.hpp>

Inheritance diagram for cv::detail::BundleAdjusterBase:

Collaboration diagram for cv::detail::BundleAdjusterBase:

Public Member Functions
double	confThresh () const
bool	operator() (const std::vector< ImageFeatures > &features, const std::vector< MatchesInfo > &pairwise_matches, std::vector< CameraParams > &cameras)
	Estimates camera parameters. More...
const Mat	refinementMask () const
void	setConfThresh (double conf_thresh)
void	setRefinementMask (const Mat &mask)
void	setTermCriteria (const TermCriteria &term_criteria)
TermCriteria	termCriteria ()

Protected Member Functions
	BundleAdjusterBase (int num_params_per_cam, int num_errs_per_measurement)
	Construct a bundle adjuster base instance. More...
virtual void	calcError (Mat &err)=0
	Calculates error vector. More...
virtual void	calcJacobian (Mat &jac)=0
	Calculates the cost function jacobian. More...
virtual bool	estimate (const std::vector< ImageFeatures > &features, const std::vector< MatchesInfo > &pairwise_matches, std::vector< CameraParams > &cameras) CV_OVERRIDE
	This method must implement camera parameters estimation logic in order to make the wrapper detail::Estimator::operator()_ work. More...
virtual void	obtainRefinedCameraParams (std::vector< CameraParams > &cameras) const =0
	Gets the refined camera parameters. More...
virtual void	setUpInitialCameraParams (const std::vector< CameraParams > &cameras)=0
	Sets initial camera parameter to refine. More...

Protected Attributes
Mat	cam_params_
double	conf_thresh_
std::vector< std::pair< int, int > >	edges_
const ImageFeatures *	features_
int	num_errs_per_measurement_
int	num_images_
int	num_params_per_cam_
const MatchesInfo *	pairwise_matches_
Mat	refinement_mask_
TermCriteria	term_criteria_
int	total_num_matches_

Detailed Description

Base class for all camera parameters refinement methods.

Constructor & Destructor Documentation

BundleAdjusterBase()

cv::detail::BundleAdjusterBase::BundleAdjusterBase ( int  num_params_per_cam, int  num_errs_per_measurement  )

inlineprotected

Construct a bundle adjuster base instance.

Parameters

num_params_per_cam	Number of parameters per camera
num_errs_per_measurement	Number of error terms (components) per match

References cv::TermCriteria::COUNT, CV_8U, CV_OVERRIDE, cv::TermCriteria::EPS, and cv::Mat::ones().

Here is the call graph for this function:

Member Function Documentation

calcError()

virtual void cv::detail::BundleAdjusterBase::calcError ( Mat &  err )

protectedpure virtual

Calculates error vector.

Parameters

err	Error column-vector of length total_num_matches * num_errs_per_measurement

calcJacobian()

virtual void cv::detail::BundleAdjusterBase::calcJacobian ( Mat &  jac )

protectedpure virtual

Calculates the cost function jacobian.

Parameters

jac	Jacobian matrix of dimensions (total_num_matches * num_errs_per_measurement) x (num_images * num_params_per_cam)

confThresh()

double cv::detail::BundleAdjusterBase::confThresh ( ) const

inline

estimate()

virtual bool cv::detail::BundleAdjusterBase::estimate ( const std::vector< ImageFeatures > &  features, const std::vector< MatchesInfo > &  pairwise_matches, std::vector< CameraParams > &  cameras  )

protectedvirtual

This method must implement camera parameters estimation logic in order to make the wrapper detail::Estimator::operator()_ work.

Parameters

features	Features of images
pairwise_matches	Pairwise matches of images
cameras	Estimated camera parameters

Returns

True in case of success, false otherwise

Implements cv::detail::Estimator.

obtainRefinedCameraParams()

virtual void cv::detail::BundleAdjusterBase::obtainRefinedCameraParams ( std::vector< CameraParams > &  cameras ) const

protectedpure virtual

Gets the refined camera parameters.

Parameters

cameras

Refined camera parameters

operator()()

bool cv::detail::Estimator::operator() ( const std::vector< ImageFeatures > &  features, const std::vector< MatchesInfo > &  pairwise_matches, std::vector< CameraParams > &  cameras  )

inlineinherited

Estimates camera parameters.

Parameters

features	Features of images
pairwise_matches	Pairwise matches of images
cameras	Estimated camera parameters

Returns

True in case of success, false otherwise

References CV_OUT.

refinementMask()

const Mat cv::detail::BundleAdjusterBase::refinementMask ( ) const

inline

References cv::Mat::clone().

Here is the call graph for this function:

setConfThresh()

void cv::detail::BundleAdjusterBase::setConfThresh ( double  conf_thresh )

inline

setRefinementMask()

void cv::detail::BundleAdjusterBase::setRefinementMask ( const Mat &  mask )

inline

References cv::Mat::clone(), CV_8U, CV_Assert, cv::Mat::size, and cv::Mat::type().

Here is the call graph for this function:

setTermCriteria()

void cv::detail::BundleAdjusterBase::setTermCriteria ( const TermCriteria &  term_criteria )

inline

setUpInitialCameraParams()

virtual void cv::detail::BundleAdjusterBase::setUpInitialCameraParams ( const std::vector< CameraParams > &  cameras )

protectedpure virtual

Sets initial camera parameter to refine.

Parameters

cameras

Camera parameters

termCriteria()

TermCriteria cv::detail::BundleAdjusterBase::termCriteria ( )

inline

Member Data Documentation

cam_params_

Mat cv::detail::BundleAdjusterBase::cam_params_

protected

conf_thresh_

double cv::detail::BundleAdjusterBase::conf_thresh_

protected

edges_

std::vector<std::pair<int,int> > cv::detail::BundleAdjusterBase::edges_

protected

features_

const ImageFeatures* cv::detail::BundleAdjusterBase::features_

protected

num_errs_per_measurement_

int cv::detail::BundleAdjusterBase::num_errs_per_measurement_

protected

num_images_

int cv::detail::BundleAdjusterBase::num_images_

protected

num_params_per_cam_

int cv::detail::BundleAdjusterBase::num_params_per_cam_

protected

pairwise_matches_

const MatchesInfo* cv::detail::BundleAdjusterBase::pairwise_matches_

protected

refinement_mask_

Mat cv::detail::BundleAdjusterBase::refinement_mask_

protected

term_criteria_

TermCriteria cv::detail::BundleAdjusterBase::term_criteria_

protected

total_num_matches_

int cv::detail::BundleAdjusterBase::total_num_matches_

protected

---

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

• opencv2/stitching/detail/motion_estimators.hpp