OpenCV  4.1.0
Open Source Computer Vision
Classes | Public Member Functions | Static Public Member Functions | List of all members
cv::TrackerMIL Class Reference

The MIL algorithm trains a classifier in an online manner to separate the object from the background. More...

#include <opencv2/tracking/tracker.hpp>

Inheritance diagram for cv::TrackerMIL:
cv::Tracker cv::Algorithm


struct  Params

Public Member Functions

virtual ~TrackerMIL () CV_OVERRIDE
- Public Member Functions inherited from cv::Tracker
virtual ~Tracker () CV_OVERRIDE
bool init (InputArray image, const Rect2d &boundingBox)
 Initialize the tracker with a known bounding box that surrounded the target.
virtual void read (const FileNode &fn) CV_OVERRIDE=0
 Reads algorithm parameters from a file storage.
bool update (InputArray image, Rect2d &boundingBox)
 Update the tracker, find the new most likely bounding box for the target.
virtual void write (FileStorage &fs) const CV_OVERRIDE=0
 Stores algorithm parameters in a file storage.
- Public Member Functions inherited from cv::Algorithm
 Algorithm ()
virtual ~Algorithm ()
virtual void clear ()
 Clears the algorithm state.
virtual bool empty () const
 Returns true if the Algorithm is empty (e.g. in the very beginning or after unsuccessful read.
virtual String getDefaultName () const
virtual void save (const String &filename) const
void write (const Ptr< FileStorage > &fs, const String &name=String()) const
 simplified API for language bindingsThis is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Static Public Member Functions

static Ptr< TrackerMILcreate (const TrackerMIL::Params &parameters)
static Ptr< TrackerMILcreate ()

Additional Inherited Members

- Protected Member Functions inherited from cv::Tracker
virtual bool initImpl (const Mat &image, const Rect2d &boundingBox)=0
virtual bool updateImpl (const Mat &image, Rect2d &boundingBox)=0
- Protected Attributes inherited from cv::Tracker
Ptr< TrackerFeatureSetfeatureSet
bool isInit
Ptr< TrackerModelmodel
Ptr< TrackerSamplersampler

Detailed Description

The MIL algorithm trains a classifier in an online manner to separate the object from the background.

Multiple Instance Learning avoids the drift problem for a robust tracking. The implementation is based onMIL .

Original code can be found here

Constructor & Destructor Documentation

virtual cv::TrackerMIL::~TrackerMIL ( )

Member Function Documentation

static Ptr<TrackerMIL> cv::TrackerMIL::create ( const TrackerMIL::Params parameters)


parametersMIL parameters TrackerMIL::Params
static Ptr<TrackerMIL> cv::TrackerMIL::create ( )

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