KLT Feature Tracker

Runs KLT Feature tracking on a sequence of frames. More...

Data Structures
struct	VPIKLTFeatureTrackerCreationParams
	Creation parameters of KLT Feature Tracker. More...
struct	VPIKLTFeatureTrackerParams
	Structure that defines the parameters for vpiCreateKLTFeatureTracker. More...
struct	VPIKLTTrackedBoundingBox
	Stores a bounding box that is being tracked by KLT Tracker. More...

Enumerations
enum	VPIKLTFeatureTrackerType
	KLT Feature Tracker algorithm type. More...

Functions
VPIStatus	vpiInitKLTFeatureTrackerCreationParams (VPIKLTFeatureTrackerCreationParams *params)
	Initialize VPIKLTFeatureTrackerCreationParams with default values. More...
VPIStatus	vpiCreateKLTFeatureTracker (uint64_t backends, int32_t imageWidth, int32_t imageHeight, VPIImageFormat imageFormat, const VPIKLTFeatureTrackerCreationParams *params, VPIPayload *payload)
	Creates payload for vpiSubmitKLTFeatureTracker. More...
VPIStatus	vpiInitKLTFeatureTrackerParams (VPIKLTFeatureTrackerParams *params)
	Initialize VPIKLTFeatureTrackerParams with default values. More...
VPIStatus	vpiSubmitKLTFeatureTracker (VPIStream stream, uint64_t backend, VPIPayload payload, VPIImage templateImage, VPIArray inputBoxList, VPIArray inputPredictionList, VPIImage referenceImage, VPIArray outputBoxList, VPIArray outputEstimationList, const VPIKLTFeatureTrackerParams *params)
	Runs KLT Feature Tracker on two frames. More...

Detailed Description

Runs KLT Feature tracking on a sequence of frames.

Refer to KLT Feature Tracker for more details and usage examples.

---

Data Structure Documentation

VPIKLTFeatureTrackerCreationParams

struct VPIKLTFeatureTrackerCreationParams

Creation parameters of KLT Feature Tracker.

Definition at line 80 of file KLTFeatureTracker.h.

Collaboration diagram for VPIKLTFeatureTrackerCreationParams:

Data Fields
int32_t	maxTemplateCount	Maximum number of templates to be tracked. Must be >= 1 and <= 128.
int32_t	maxTemplateWidth	Maximum width of each tracked template. It must be >= 4 and <= 128 (on PVA, <= 64).
int32_t	maxTemplateHeight	Maximum height of each tracked templates. It must be >= 4 and <= 128 (on PVA, <= 64).

VPIKLTFeatureTrackerParams

struct VPIKLTFeatureTrackerParams

Structure that defines the parameters for vpiCreateKLTFeatureTracker.

Definition at line 173 of file KLTFeatureTracker.h.

Collaboration diagram for VPIKLTFeatureTrackerParams:

Data Fields
int32_t	numberOfIterationsScaling	Number of Inverse compositional iterations of scale estimations. On PVA the maximum allowed is 20.
float	nccThresholdUpdate	Threshold for requiring template update. Must be a value between 0 and 1. Must be greater than nccThresholdKill.
float	nccThresholdKill	Threshold to consider template tracking was lost. Must be a value between 0 and 1.
float	nccThresholdStop	Threshold to early stop iteration. Must be a value between 0 and 1. Must be greater than nccThresholdUpdate.
float	maxScaleChange	Maximum relative scale change. Scale changes larger than this will make KLT consider that tracking was lost. Must be >= 0. On PVA, maximum scale change is 0.2.
float	maxTranslationChange	Maximum relative translation change. Must be less >= 0. Translation changes larger than this will make KLT consider that tracking was lost.
VPIKLTFeatureTrackerType	trackingType	Type of KLT tracking that will be performed.

VPIKLTTrackedBoundingBox

struct VPIKLTTrackedBoundingBox

Stores a bounding box that is being tracked by KLT Tracker.

Definition at line 491 of file Types.h.

Collaboration diagram for VPIKLTTrackedBoundingBox:

Data Fields
VPIBoundingBox	bbox	Bounding box being tracked.
int8_t	trackingStatus	Tracking status of this bounding box. Accepted values: 1 tracking information is invalid and shouldn't be relied upon. 0 tracking information is valid.
int8_t	templateStatus	Status of the template related to this bounding box. Accepted values: 1 template needs updating. 0 existing template still can be used for tracking, it doesn't need to be updated.
uint8_t	reserved1	Reserved for future use.
uint8_t	reserved2	Reserved for future use.

Enumeration Type Documentation

VPIKLTFeatureTrackerType

enum VPIKLTFeatureTrackerType

#include <vpi/algo/KLTFeatureTracker.h>

KLT Feature Tracker algorithm type.

Enumerator
VPI_KLT_INVERSE_COMPOSITIONAL	Inverse compositional algorithm for KLT tracker. The inverse compositional algorithm is a reformulation of the classic Lucas-Kanade algorithm to make the steepest-descent images and Hessian constant. Ref: Simon Baker, Iain Matthew, "Lucas-Kanade 20 Years On: A Unified Framework". International Journal of Computer Vision, February 2004, Volume 56, issue 3, pp. 228-231.

Definition at line 156 of file KLTFeatureTracker.h.

Function Documentation

vpiInitKLTFeatureTrackerCreationParams()

VPIStatus vpiInitKLTFeatureTrackerCreationParams

(

VPIKLTFeatureTrackerCreationParams *

params

)

#include <vpi/algo/KLTFeatureTracker.h>

Initialize VPIKLTFeatureTrackerCreationParams with default values.

Default values:

• maxTemplateCount: 128
• maxTemplateWidth: 64
• maxTemplateHeight: 64

Parameters

[out]

params

Structure to be filled with default values.

Return values

VPI_ERROR_INVALID_ARGUMENT	params is NULL.
VPI_SUCCESS	Operation executed successfully.

vpiCreateKLTFeatureTracker()

VPIStatus vpiCreateKLTFeatureTracker	(	uint64_t	backends,
		int32_t	imageWidth,
		int32_t	imageHeight,
		VPIImageFormat	imageFormat,
		const VPIKLTFeatureTrackerCreationParams *	params,
		VPIPayload *	payload
	)

#include <vpi/algo/KLTFeatureTracker.h>

Creates payload for vpiSubmitKLTFeatureTracker.

Parameters

[in]	backends	VPI backends that are eligible to execute the algorithm. Valid values: VPI_BACKEND_CPU VPI_BACKEND_CUDA VPI_BACKEND_PVA Backend must be enabled in current context.
[in]	imageWidth,imageHeight	Input image dimensions. On PVA, dimensions must be between 65x65 and 3264x2448. Must be >= 0 (other backends).
[in]	imageFormat	Input image format. The accepted image formats are: VPI_IMAGE_FORMAT_U8 VPI_IMAGE_FORMAT_S8 VPI_IMAGE_FORMAT_U16 (PVA only supports this format, pixel values' range must be between 0 and 255) VPI_IMAGE_FORMAT_S16
[in]	params	params
[out]	payload	Pointer to the payload variable that receives the created handle.

Return values

VPI_ERROR_INVALID_ARGUMENT	payload is NULL.
VPI_ERROR_INVALID_ARGUMENT	imageWidth and imageHeight outside valid range.
VPI_ERROR_INVALID_ARGUMENT	maxTemplateWidth and maxTemplateHeight in params outside valid range.
VPI_ERROR_INVALID_ARGUMENT	maxTemplateCount in params outside valid range.
VPI_ERROR_INVALID_ARGUMENT	backends refers to an invalid backend.
VPI_IMAGE_FORMAT_INVALID	imageFormat is not supported.
VPI_ERROR_INVALID_OPERATION	PVA hardware is not available.
VPI_ERROR_INVALID_OPERATION	Backend isn't enabled in current context.
VPI_ERROR_NOT_IMPLEMENTED	KLT Feature Tracker algorithm is not supported by given backend.
VPI_ERROR_INVALID_CONTEXT	Current context is destroyed.
VPI_ERROR_OUT_OF_MEMORY	Cannot allocate required resources.
VPI_SUCCESS	Operation executed successfully.

vpiInitKLTFeatureTrackerParams()

VPIStatus vpiInitKLTFeatureTrackerParams

(

VPIKLTFeatureTrackerParams *

params

)

#include <vpi/algo/KLTFeatureTracker.h>

Initialize VPIKLTFeatureTrackerParams with default values.

Default values:

• numberOfIterationsScaling: 20
• nccThresholdUpdate: 0.8
• nccThresholdKill: 0.6
• nccThresholdStop: 1.0
• maxScaleChange: 0.2
• maxTranslationChange: 1.5
• trackingType: VPI_KLT_INVERSE_COMPOSITIONAL

Parameters

[out]

params

Structure to be filled with default values.

Returns

an error code on failure else VPI_SUCCESS.

vpiSubmitKLTFeatureTracker()

VPIStatus vpiSubmitKLTFeatureTracker	(	VPIStream	stream,
		uint64_t	backend,
		VPIPayload	payload,
		VPIImage	templateImage,
		VPIArray	inputBoxList,
		VPIArray	inputPredictionList,
		VPIImage	referenceImage,
		VPIArray	outputBoxList,
		VPIArray	outputEstimationList,
		const VPIKLTFeatureTrackerParams *	params
	)

#include <vpi/algo/KLTFeatureTracker.h>

Runs KLT Feature Tracker on two frames.

Outputs tracked bounding boxes and estimated transform array.

Parameters

[in]	stream	The stream where the operation will be queued in. Must not be NULL. Stream must have enabled the backends that will execute the algorithm.
[in]	backend	Backend that will execute the algorithm. Must be the backend specified during payload creation or 0 as a shorthand to use this backend.
[in]	payload	Payload created by vpiCreateKLTFeatureTracker.
[in]	referenceImage	Reference image. Must not be NULL. Must have same dimensions and format as template image. Image must have enabled the backends that will execute the algorithm.
[in]	inputBoxList	Input bounding box array. Must not be NULL. Must have type VPI_ARRAY_TYPE_KLT_TRACKED_BOUNDING_BOX. Array must have enabled the backends that will execute the algorithm. On PVA, array capacity must be >= 128.
[in]	inputPredictionList	Input predicted transform array. Must not be NULL. Must have type VPI_ARRAY_TYPE_HOMOGRAPHY_TRANSFORM_2D. Array must have enabled the backends that will execute the algorithm. On PVA, array capacity must be >= 128.
[in]	templateImage	Template image. Must not be NULL. Must have same dimensions and format as referenceImage. Image must have enabled the backends that will execute the algorithm.
[out]	outputBoxList	Output Bounding box array. Must not be NULL. Must have type VPI_ARRAY_TYPE_KLT_TRACKED_BOUNDING_BOX. Array must have enabled the backends that will execute the algorithm. On PVA, array capacity must be >= 128.
[out]	outputEstimationList	Estimated transform array. Must not be NULL. Must have type VPI_ARRAY_TYPE_HOMOGRAPHY_TRANSFORM_2D. Array must have enabled the backends that will execute the algorithm. On PVA, array capacity must be >= 128.
[in]	params	Control parameters of the KLT feature tracker algorithm.

Return values

VPI_ERROR_INVALID_ARGUMENT	stream is NULL.
VPI_ERROR_INVALID_ARGUMENT	referenceImage, templateImage, inputBoxList, inputPredictionList, outputBoxList or outputEstimationList are NULL.
VPI_ERROR_INVALID_ARGUMENT	params is NULL.
VPI_ERROR_INVALID_ARGUMENT	payload not created by vpiCreateKLTFeatureTracker.
VPI_ERROR_INVALID_ARGUMENT	templateImage and referenceImage dimensions and format do not match.
VPI_ERROR_INVALID_ARGUMENT	Input and parameter image dimension do not match.
VPI_ERROR_INVALID_ARGUMENT	Image format and payload image format do not match.
VPI_ERROR_INVALID_ARGUMENT	Unsupported array type for inputBoxList or outputBoxList.
VPI_ERROR_INVALID_ARGUMENT	Unsupported array type for inputPredictionList or outputEstimationList.
VPI_ERROR_INVALID_ARGUMENT	Parameter(s) in params outside valid range.
VPI_ERROR_INVALID_PAYLOAD_TYPE	payload is invalid.
VPI_ERROR_INVALID_OPERATION	The needed backends aren't enabled in stream, referenceImage, templateImage, inputBoxList, inputPredictionList, outputBoxList or outputEstimationList.
VPI_SUCCESS	Operation executed successfully.