vpRBDriftDetector Class Reference

#include <vpRBDriftDetector.h>

Inheritance diagram for vpRBDriftDetector:

Public Member Functions
	vpRBDriftDetector ()=default
virtual	~vpRBDriftDetector ()=default
virtual void	update (const vpRBFeatureTrackerInput &previousFrame, const vpRBFeatureTrackerInput &frame, const vpHomogeneousMatrix &cTo, const vpHomogeneousMatrix &cprevTo)=0
virtual double	score (const vpRBFeatureTrackerInput &frame, const vpHomogeneousMatrix &cTo)=0
virtual double	getScore () const =0
virtual bool	hasDiverged () const =0
virtual void	reset ()
virtual void	display (const vpImage< vpRGBa > &I)=0
virtual void	loadJsonConfiguration (const nlohmann::json &)=0

Detailed Description

Base interface for algorithms that should detect tracking drift for the render-based tracker.

In the tracking loop, these algorithms should be used as follows:

• Perform a tracking step, estimating a new object pose
• Call vpRBDriftDetector::update to update the drift detection parameters.
• use vpRBDriftDetector::hasDiverged to detect the drift, or vpRBDriftDetector::getScore to use the estimated tracking reliability.

Tutorials & Examples

Tutorials
If you want to have an in-depth presentation of the Render-Based Tracker (RBT), you may have a look at:

• Tutorial: Tracking with the Render-Based Tracker

Definition at line 68 of file vpRBDriftDetector.h.

Constructor & Destructor Documentation

vpRBDriftDetector()

vpRBDriftDetector::vpRBDriftDetector ( )

default

~vpRBDriftDetector()

virtual vpRBDriftDetector::~vpRBDriftDetector ( )

virtualdefault

Member Function Documentation

display()

virtual void vpRBDriftDetector::display ( const vpImage< vpRGBa > & I )

pure virtual

Displays the information used for drift detection.

Parameters

I	the image in which to display the information

Implemented in vpRBProbabilistic3DDriftDetector.

getScore()

virtual double vpRBDriftDetector::getScore ( ) const

pure virtual

Get the estimated tracking reliability. A high score should mean that the tracking is reliable. Different algorithms may use different value ranges.

Returns

The estimated tracking accuracy

Implemented in vpRBProbabilistic3DDriftDetector.

hasDiverged()

virtual bool vpRBDriftDetector::hasDiverged ( ) const

pure virtual

Returns whether the tracking has diverged and should be reinitialized. This function should be called after update.

Implemented in vpRBProbabilistic3DDriftDetector.

loadJsonConfiguration()

virtual void vpRBDriftDetector::loadJsonConfiguration ( const nlohmann::json & )

pure virtual

Implemented in vpRBProbabilistic3DDriftDetector.

reset()

virtual void vpRBDriftDetector::reset ( )

inlinevirtual

Reimplemented in vpRBProbabilistic3DDriftDetector.

Definition at line 103 of file vpRBDriftDetector.h.

score()

virtual double vpRBDriftDetector::score ( const vpRBFeatureTrackerInput & frame, const vpHomogeneousMatrix & cTo )

pure virtual

Implemented in vpRBProbabilistic3DDriftDetector.

update()

virtual void vpRBDriftDetector::update ( const vpRBFeatureTrackerInput & previousFrame, const vpRBFeatureTrackerInput & frame, const vpHomogeneousMatrix & cTo, const vpHomogeneousMatrix & cprevTo )

pure virtual

Update the algorithm after a new tracking step.

Parameters

previousFrame	The previous frame data: contains the input images at t-1 (linked to cprevTo) and the renders at t-2. May be empty for the first iteration
frame	The current frame data: contains the input images at time t (linked to the newly estimated cTo) and the renders at t-1 (linked to cprevTo)
cTo	the newly estimated object pose in the camera frame
cprevTo	the previously estimated object pose in the camera frame

Implemented in vpRBProbabilistic3DDriftDetector.