opentl::modalities::ContourPointsGPU Class Reference

Class for contour point feature modality computation. ATTENTION: This class is currently not prepared for multi-threaded match...Level() calls related to a single sensor index! More...

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).
	ContourPointsGPU (const ContourPointsGPU &c)
	Deep Copy Constructor.
	ContourPointsGPU (const std::vector< opentl::models::ObjModel * > &objModels, opentl::modelprojection::Warp *warp, opentl::modelprojection::GLScene *scene, int camIdx, modelprojection::VisibilityCheck *visCheck=NULL)
	Constructor ATTENTION: The sensor vector element size/sequence must be identical for the warps of all object models !!!
opentl::core::cvdata::VisContourSamplePts *	getContourData (int s=0)
virtual void	init ()
	Allocate data structures required by this class.
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)
	Matching on feature level match projected contour points with detected image edges.
virtual int	preProcess (const opentl::core::cvdata::Image &image)
	Compute the edge map of I (e.g. with Canny), and store it into mImageFeatDetected SensorData input image has to be a GRAYSCALE image.
virtual int	sampleModelFeatures (const TargetPtrVector &targets)
	Do GPU-based visible edge sample point computation.
void	setMeasStd (double std)
virtual	~ContourPointsGPU ()
	Destructor.
Public Attributes
opentl::core::cvdata::Image *	mImageDebugOutput
	Debug Image output.

---

Detailed Description

Class for contour point feature modality computation. ATTENTION: This class is currently not prepared for multi-threaded match...Level() calls related to a single sensor index!

Author:

Erwin Roth <eroth@mytum.de>

---

Member Enumeration Documentation

enum opentl::modalities::ContourPointsGPU::FeatDetectFilterType

Possible filter types for image feature detection.

Enumerator:

CANNY
SOBEL
CANNY_AND_SOBEL

enum opentl::modalities::ContourPointsGPU::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)
construct_loadShaderCodeFromFile	(bool) Load Shader code from File
construct_sampleSilhouetteEdgesOnly	(bool) Sample silhoutte edged only
construct_MaxSamplePointsPerTarget	(int) maximum sample points per target
preProcess_featureFilterType	(FeatDetectFilterType) preprocess filter type
OFFLINE_COUNT

Reimplemented from opentl::modalities::Modality.

enum opentl::modalities::ContourPointsGPU::OnlineParameters

Enumerator:

sample_discardBackgroundEdgeWithinPixelRange	(unsigned int) range to discard background edge pixel
sample_computeScreenPointJacobians	(bool) compute screen point jacobians
preProcess_cannyThreshold1	(double) canny threshold1
preProcess_cannyThreshold2	(double) canny threshold2
matchF_edgeSearchlengthAlongNormalinPixel	(unsigned int) edge search length along the normal in pixel
matchF_missingAssocRate	(double)
matchF_useRansac	(bool) use RANSAC
matchF_dotprodThreshold	(double) dot product threshold
matchF_angleThreshold	(double) angle threshold
matchF_enableDebugOutput	(bool) enable debug output
matchFeatLevel_robustFlag	(bool) robust falg
matchFeatLevel_useFixedCov	(bool) use fixed covariance
matchFeatLevel_fixedCov	(double) ficed covariance
matchF_nearestNeighbor	(bool) nearest neighbour
ONLINE_COUNT

Reimplemented from opentl::modalities::Modality.

---

Constructor & Destructor Documentation

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

Constructor ATTENTION: The sensor vector element size/sequence must be identical for the warps of all object models !!!

opentl::modalities::ContourPointsGPU::ContourPointsGPU

(

const ContourPointsGPU &

c

)

Deep Copy Constructor.

virtual opentl::modalities::ContourPointsGPU::~ContourPointsGPU

(

)

[virtual]

Destructor.

---

Member Function Documentation

virtual modalities::Modality* opentl::modalities::ContourPointsGPU::clone

(

)

const [virtual]

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

Implements opentl::modalities::Modality.

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

(

int

s = 0

)

[inline]

virtual void opentl::modalities::ContourPointsGPU::init

(

)

[virtual]

Allocate data structures required by this class.

Reimplemented from opentl::modalities::Modality.

virtual void opentl::modalities::ContourPointsGPU::invalidateFeatures

(

)

[virtual]

Set the mUpToDate flag in the feature data to FALSE.

Reimplemented from opentl::modalities::Modality.

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

Matching on feature level

• match projected contour points with detected image edges.

Parameters:

	targets	Vector of targets
	outputMeas	Feature-space measurements and residuals (one per target and internal state)

Reimplemented from opentl::modalities::Modality.

virtual int opentl::modalities::ContourPointsGPU::preProcess

(

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

image

)

[virtual]

Compute the edge map of I (e.g. with Canny), and store it into mImageFeatDetected SensorData input image has to be a GRAYSCALE image.

virtual int opentl::modalities::ContourPointsGPU::sampleModelFeatures

(

const TargetPtrVector &

targets

)

[virtual]

Do GPU-based visible edge sample point computation.

Parameters:

targets

Vector of targets

Reimplemented from opentl::modalities::Modality.

void opentl::modalities::ContourPointsGPU::setMeasStd

(

double

std

)

[inline]

---

Member Data Documentation

opentl::core::cvdata::Image* opentl::modalities::ContourPointsGPU::mImageDebugOutput

Debug Image output.