Loading...
Searching...
No Matches

Abstract base class for TrackerStateEstimator that estimates the most likely target state. More...

#include <opencv2/video/detail/tracking.detail.hpp>

Inheritance diagram for cv::detail::tracking::TrackerStateEstimator:
cv::detail::tracking::TrackerStateEstimatorAdaBoosting cv::detail::tracking::TrackerStateEstimatorSVM

Public Member Functions

virtual ~TrackerStateEstimator ()
 
Ptr< TrackerTargetStateestimate (const std::vector< ConfidenceMap > &confidenceMaps)
 Estimate the most likely target state, return the estimated state.
 
String getClassName () const
 Get the name of the specific TrackerStateEstimator.
 
void update (std::vector< ConfidenceMap > &confidenceMaps)
 Update the ConfidenceMap with the scores.
 

Static Public Member Functions

static Ptr< TrackerStateEstimatorcreate (const String &trackeStateEstimatorType)
 Create TrackerStateEstimator by tracker state estimator type.
 

Protected Member Functions

virtual Ptr< TrackerTargetStateestimateImpl (const std::vector< ConfidenceMap > &confidenceMaps)=0
 
virtual void updateImpl (std::vector< ConfidenceMap > &confidenceMaps)=0
 

Protected Attributes

String className
 

Detailed Description

Abstract base class for TrackerStateEstimator that estimates the most likely target state.

See [AAM] State estimator

See [AMVOT] Statistical modeling (Fig. 3), Table III (generative) - IV (discriminative) - V (hybrid)

Constructor & Destructor Documentation

◆ ~TrackerStateEstimator()

virtual cv::detail::tracking::TrackerStateEstimator::~TrackerStateEstimator ( )
virtual

Member Function Documentation

◆ create()

static Ptr< TrackerStateEstimator > cv::detail::tracking::TrackerStateEstimator::create ( const String trackeStateEstimatorType)
static

Create TrackerStateEstimator by tracker state estimator type.

Parameters
trackeStateEstimatorTypeThe TrackerStateEstimator name

The modes available now:

  • "BOOSTING" – Boosting-based discriminative appearance models. See [AMVOT] section 4.4

The modes available soon:

  • "SVM" – SVM-based discriminative appearance models. See [AMVOT] section 4.5

◆ estimate()

Ptr< TrackerTargetState > cv::detail::tracking::TrackerStateEstimator::estimate ( const std::vector< ConfidenceMap > &  confidenceMaps)

Estimate the most likely target state, return the estimated state.

Parameters
confidenceMapsThe overall appearance model as a list of :cConfidenceMap

◆ estimateImpl()

virtual Ptr< TrackerTargetState > cv::detail::tracking::TrackerStateEstimator::estimateImpl ( const std::vector< ConfidenceMap > &  confidenceMaps)
protectedpure virtual

◆ getClassName()

String cv::detail::tracking::TrackerStateEstimator::getClassName ( ) const

Get the name of the specific TrackerStateEstimator.

◆ update()

void cv::detail::tracking::TrackerStateEstimator::update ( std::vector< ConfidenceMap > &  confidenceMaps)

Update the ConfidenceMap with the scores.

Parameters
confidenceMapsThe overall appearance model as a list of :cConfidenceMap

◆ updateImpl()

virtual void cv::detail::tracking::TrackerStateEstimator::updateImpl ( std::vector< ConfidenceMap > &  confidenceMaps)
protectedpure virtual

Member Data Documentation

◆ className

String cv::detail::tracking::TrackerStateEstimator::className
protected

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