An Opaque object that contain a classifier model. The model can be cascade model or SVM model or any other machine learning model. More...

Typedefs
typedef struct _vx_classifier_model *	vx_classifier_model
	classification model to be used in `vxScanClassifierNode`. The classification models are loadable by undefined binary format see `vxImportClassifierModel`. Extensions will be added to the specification, to support a defined binary format.

Enumerations
enum	vx_classifier_model_format_e { VX_CLASSIFIER_MODEL_UNDEFINED = ((( VX_ID_KHRONOS ) << 20) \| ( VX_ENUM_CLASSIFIER_MODEL << 12)) + 0x0 }
	Classifier model format enums. In the main specification only undefined binary format is supported. Extensions to the specification will be added in order to support specific binary format. More...

Functions
vx_classifier_model	vxImportClassifierModel (vx_context context, vx_enum format, vx_uint8 *ptr, vx_size length)
	Creates an opaque reference classifier model This function creates a classifier model to be used in `vxScanClassifierNode`. The object classifier object is a read-only constant object. It cannot be changed during graph execution. More...

vx_status	vxReleaseClassifierModel (vx_classifier_model *model)
	Releases a reference of an ClassifierModel object. The object may not be garbage collected until its total reference and its contained objects count is zero. After returning from this function the reference is zeroed/cleared. More...

Detailed Description

An Opaque object that contain a classifier model. The model can be cascade model or SVM model or any other machine learning model.

The Object is created by importing data from a binary format. The specification will not define such a format. Extensions to the specification will be added in order to define such binary formats.

Enumeration Type Documentation

§ vx_classifier_model_format_e

enum vx_classifier_model_format_e

Classifier model format enums. In the main specification only undefined binary format is supported. Extensions to the specification will be added in order to support specific binary format.

Enumerator
VX_CLASSIFIER_MODEL_UNDEFINED	Undefined binary format. Using this enumeration will result in an implementation defined behaviour.

Definition at line 80 of file vx_khr_class.h.

Function Documentation

§ vxImportClassifierModel()

vx_classifier_model vxImportClassifierModel	(	vx_context	context,
		vx_enum	format,
		vx_uint8 *	ptr,
		vx_size	length
	)

Creates an opaque reference classifier model This function creates a classifier model to be used in vxScanClassifierNode. The object classifier object is a read-only constant object. It cannot be changed during graph execution.

Parameters

[in]	context	Reference to the context where to create the ClassifierModel.
[in]	format	The binary format which contain the classifier model. See `vx_classifier_model_format_e`. Currently only undefined binary format is supported. Extensions will be added to the specification, to support a classification model defined binary format.
[in]	ptr	A memory pointer to the binary format.
[in]	length	size in bytes of binary format data.

Returns: A ClassifierModel reference vx_classifier_model. Any possible errors preventing a successful creation should be checked using vxGetStatus.

§ vxReleaseClassifierModel()

vx_status vxReleaseClassifierModel ( vx_classifier_model * model )

Releases a reference of an ClassifierModel object. The object may not be garbage collected until its total reference and its contained objects count is zero. After returning from this function the reference is zeroed/cleared.

Parameters

[in] model The pointer to the ClassifierModel to release.

Returns: A vx_status_e enumeration.

Return values

<tt>

Typedefs