OpenCV  4.1.0
Open Source Computer Vision
Public Member Functions | Protected Member Functions | Protected Attributes | List of all members
cv::TrackerModel Class Referenceabstract

Abstract class that represents the model of the target. It must be instantiated by specialized tracker. More...

#include <opencv2/tracking/tracker.hpp>

Public Member Functions

 TrackerModel ()
 Constructor.
 
virtual ~TrackerModel ()
 Destructor.
 
const std::vector
< ConfidenceMap > & 
getConfidenceMaps () const
 Get the list of the ConfidenceMap.
 
const ConfidenceMapgetLastConfidenceMap () const
 Get the last ConfidenceMap for the current frame.
 
Ptr< TrackerTargetStategetLastTargetState () const
 Get the last TrackerTargetState from Trajectory.
 
Ptr< TrackerStateEstimatorgetTrackerStateEstimator () const
 Get the TrackerStateEstimator.
 
void modelEstimation (const std::vector< Mat > &responses)
 Estimate the most likely target location.
 
void modelUpdate ()
 Update the model.
 
bool runStateEstimator ()
 Run the TrackerStateEstimator, return true if is possible to estimate a new state, false otherwise.
 
void setLastTargetState (const Ptr< TrackerTargetState > &lastTargetState)
 Set the current TrackerTargetState in the Trajectory.
 
bool setTrackerStateEstimator (Ptr< TrackerStateEstimator > trackerStateEstimator)
 Set TrackerEstimator, return true if the tracker state estimator is added, false otherwise.
 

Protected Member Functions

virtual void modelEstimationImpl (const std::vector< Mat > &responses)=0
 
virtual void modelUpdateImpl ()=0
 

Protected Attributes

std::vector< ConfidenceMapconfidenceMaps
 
ConfidenceMap currentConfidenceMap
 
int maxCMLength
 
Ptr< TrackerStateEstimatorstateEstimator
 
Trajectory trajectory
 

Detailed Description

Abstract class that represents the model of the target. It must be instantiated by specialized tracker.

SeeAAM Ak

Inherits this with your TrackerModel

Constructor & Destructor Documentation

cv::TrackerModel::TrackerModel ( )

Constructor.

virtual cv::TrackerModel::~TrackerModel ( )
virtual

Destructor.

Member Function Documentation

const std::vector<ConfidenceMap>& cv::TrackerModel::getConfidenceMaps ( ) const

Get the list of the ConfidenceMap.

const ConfidenceMap& cv::TrackerModel::getLastConfidenceMap ( ) const

Get the last ConfidenceMap for the current frame.

Ptr<TrackerTargetState> cv::TrackerModel::getLastTargetState ( ) const

Get the last TrackerTargetState from Trajectory.

Ptr<TrackerStateEstimator> cv::TrackerModel::getTrackerStateEstimator ( ) const
void cv::TrackerModel::modelEstimation ( const std::vector< Mat > &  responses)

Estimate the most likely target location.

AAM ME, Model Estimation table I

Parameters
responsesFeatures extracted from TrackerFeatureSet
virtual void cv::TrackerModel::modelEstimationImpl ( const std::vector< Mat > &  responses)
protectedpure virtual
void cv::TrackerModel::modelUpdate ( )

Update the model.

AAM MU, Model Update table I

virtual void cv::TrackerModel::modelUpdateImpl ( )
protectedpure virtual
bool cv::TrackerModel::runStateEstimator ( )

Run the TrackerStateEstimator, return true if is possible to estimate a new state, false otherwise.

void cv::TrackerModel::setLastTargetState ( const Ptr< TrackerTargetState > &  lastTargetState)

Set the current TrackerTargetState in the Trajectory.

Parameters
lastTargetStateThe current TrackerTargetState
bool cv::TrackerModel::setTrackerStateEstimator ( Ptr< TrackerStateEstimator trackerStateEstimator)

Set TrackerEstimator, return true if the tracker state estimator is added, false otherwise.

Parameters
trackerStateEstimatorThe TrackerStateEstimator
Note
You can add only one TrackerStateEstimator

Member Data Documentation

std::vector<ConfidenceMap> cv::TrackerModel::confidenceMaps
protected
ConfidenceMap cv::TrackerModel::currentConfidenceMap
protected
int cv::TrackerModel::maxCMLength
protected
Ptr<TrackerStateEstimator> cv::TrackerModel::stateEstimator
protected
Trajectory cv::TrackerModel::trajectory
protected

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