opentl::modalities::ContourCCD Class Reference

Inherits opentl::modalities::Modality.

List of all members.

Public Types
Public Member Functions
virtual modalities::Modality *	clone () const
	Clone this class and all potential childs RECURSIVELY!! (deep copy).
	ContourCCD (const ContourCCD &c)
	Deep Copy Constructor.
	ContourCCD (const std::vector< opentl::models::ObjModel * > &objModels, opentl::modelprojection::Warp *warp, opentl::modelprojection::GLScene *scene, int camIdx, modelprojection::VisibilityCheck *visCheck=NULL)
void	DrawContour (opentl::core::cvdata::Image *img, CvScalar color, int targetIdx)
int	getCamIdx ()
	returns the camera index of this class instance
opentl::core::cvdata::VisContourSamplePts *	getContourData (int s=0)
virtual void	init ()
	Initialize CCD: create (if necessary) the visibility sampler.
virtual void	invalidateFeatures ()
	Set the mUpToDate flag in the feature data to FALSE.
virtual int	matchFeatLevel (const TargetPtrVector &targets, T_MEAS_FEATPtrVector &outputMeas, std::size_t partitionIdx)
	Feature-level matching for the CCD modality. It basically computes local color statistics (=features) along the visible edges on both sides. The output measurement are the color residuals, and their Jacobian w.r.t. pose parameters.
virtual int	matchObjLevel (const TargetPtrVector &targets, T_MEAS_OBJPtrVector &outputMeas, std::size_t partitionIdx)
	(Object-level is not available in CCD, in order to avoid code duplication -> use DataFusion upgrade instead).
virtual int	preProcess (const opentl::core::cvdata::Image &image)
	Do nothing (in CCD there is no pre-processing, and D=I) SensorData input image has to be an RGB image.
virtual int	sampleModelFeatures (const TargetPtrVector &targets)
	Call the GPU-based visible edge sampler.
virtual	~ContourCCD ()
	Destructor.

---

Member Enumeration Documentation

enum opentl::modalities::ContourCCD::OfflineParameters

Common flags to all modalities: they specify which field will be filled in the output measurement, inside the matchXLevel() function. Default values = all true.

Enumerator:

generate_visCheck	(bool) If true, create the visibility algorithm internally; else, use a pre-defined one (visCheck)
contour_maxSamplePointsPerTarget	(int) Subsampling for the GPU contour sampling code
CCD_DOF	(int) Number of degrees of freedom
CCD_KMIN	(int) Minimum number of contour sample positions
CCD_KMAX	(int) Maximum number of contour sample positions
CCD_L	(int) Number of samples along the normal
OFFLINE_COUNT

Reimplemented from opentl::modalities::Modality.

enum opentl::modalities::ContourCCD::OnlineParameters

Enumerator:

space_blur	(double) Space blur: this is the extent of mixing for local color statistics along contour positions in space (exponential filtering)
max_dist	(std::vector(double)) Search distance (h) at iteration 0, for each link
dist_upd	(std::vector(double)) Update coefficient of distance (h) from step to step (exponential decay), for each link
ONLINE_COUNT

Reimplemented from opentl::modalities::Modality.

---

Constructor & Destructor Documentation

opentl::modalities::ContourCCD::ContourCCD	(	const std::vector< opentl::models::ObjModel * > &	objModels,
		opentl::modelprojection::Warp *	warp,
		opentl::modelprojection::GLScene *	scene,
		int	camIdx,
		modelprojection::VisibilityCheck *	visCheck = NULL
	)

opentl::modalities::ContourCCD::ContourCCD

(

const ContourCCD &

c

)

Deep Copy Constructor.

virtual opentl::modalities::ContourCCD::~ContourCCD

(

)

[virtual]

Destructor.

---

Member Function Documentation

virtual modalities::Modality* opentl::modalities::ContourCCD::clone

(

)

const [virtual]

Clone this class and all potential childs RECURSIVELY!! (deep copy).

Implements opentl::modalities::Modality.

void opentl::modalities::ContourCCD::DrawContour	(	opentl::core::cvdata::Image *	img,
		CvScalar	color,
		int	targetIdx
	)

int opentl::modalities::ContourCCD::getCamIdx

(

)

[inline]

returns the camera index of this class instance

Reimplemented from opentl::modalities::Modality.

opentl::core::cvdata::VisContourSamplePts* opentl::modalities::ContourCCD::getContourData

(

int

s = 0

)

[inline]

virtual void opentl::modalities::ContourCCD::init

(

)

[virtual]

Initialize CCD: create (if necessary) the visibility sampler.

Reimplemented from opentl::modalities::Modality.

virtual void opentl::modalities::ContourCCD::invalidateFeatures

(

)

[virtual]

Set the mUpToDate flag in the feature data to FALSE.

Reimplemented from opentl::modalities::Modality.

virtual int opentl::modalities::ContourCCD::matchFeatLevel	(	const TargetPtrVector &	targets,
		T_MEAS_FEATPtrVector &	outputMeas,
		std::size_t	partitionIdx
	)			[virtual]

Feature-level matching for the CCD modality. It basically computes local color statistics (=features) along the visible edges on both sides. The output measurement are the color residuals, and their Jacobian w.r.t. pose parameters.

Reimplemented from opentl::modalities::Modality.

virtual int opentl::modalities::ContourCCD::matchObjLevel	(	const TargetPtrVector &	targets,
		T_MEAS_OBJPtrVector &	outputMeas,
		std::size_t	partitionIdx
	)			[inline, virtual]

(Object-level is not available in CCD, in order to avoid code duplication -> use DataFusion upgrade instead).

Reimplemented from opentl::modalities::Modality.

virtual int opentl::modalities::ContourCCD::preProcess

(

const opentl::core::cvdata::Image &

image

)

[virtual]

Do nothing (in CCD there is no pre-processing, and D=I) SensorData input image has to be an RGB image.

virtual int opentl::modalities::ContourCCD::sampleModelFeatures

(

const TargetPtrVector &

targets

)

[virtual]

Call the GPU-based visible edge sampler.

Reimplemented from opentl::modalities::Modality.