Generates a set of default bounding boxes with different sizes and aspect ratios. More...

#include <prior_box.hpp>

Inheritance diagram for cldnn::prior_box:

Collaboration diagram for cldnn::prior_box:

[legend]

Public Types
typedef cldnn_prior_box_desc	dto

Public Member Functions
	prior_box (const primitive_id &id, const primitive_id &input, const tensor &img_size, const std::vector< float > &min_sizes, const std::vector< float > &max_sizes={}, const std::vector< float > &aspect_ratios={}, const bool flip=true, const bool clip=false, const std::vector< float > &variance={}, const float step_width=0.f, const float step_height=0.f, const float offset=0.5f, const padding &output_padding=padding())
	Constructs prior-box primitive. More...

	prior_box (const dto *dto)
	Constructs a copy from C API cldnn_prior-box_desc.

Public Member Functions inherited from cldnn::primitive_base< prior_box, cldnn_prior_box_desc >
const cldnn_primitive_desc *	get_dto () const override
	Returns pointer to a C API primitive descriptor casted to cldnn_primitive_desc.

Public Member Functions inherited from cldnn::primitive
	primitive (const primitive_type_id &type, const primitive_id &id, const std::vector< primitive_id > &input, const padding &output_padding=padding())

	primitive (const cldnn_primitive_desc *dto)
	Constructs a copy from basic C API cldnn_primitive_desc.

std::vector< std::reference_wrapper< primitive_id > >	dependecies ()
	Returns references to all primitive ids on which this primitive depends - inputs, weights, biases, etc.

std::vector< primitive_id >	dependecies () const
	Returns copy of all primitive ids on which this primitive depends - inputs, weights, biases, etc.

	operator primitive_id () const
	Implicit conversion to primiitive id.

Static Public Member Functions
static primitive_type_id	type_id ()

Public Attributes
tensor	img_size
	Image width and height.

std::vector< float >	min_sizes
	Minimum box sizes in pixels.

std::vector< float >	max_sizes
	Maximum box sizes in pixels.

std::vector< float >	aspect_ratios
	Various of aspect ratios. Duplicate ratios will be ignored.

bool	flip
	If true, will flip each aspect ratio. For example, if there is aspect ratio "r", aspect ratio "1.0/r" we will generated as well.

bool	clip
	If true, will clip the prior so that it is within [0, 1].

std::vector< float >	variance
	Variance for adjusting the prior boxes.

float	step_width
	Step width.

float	step_height
	Step height.

float	offset
	Offset to the top left corner of each cell.

Public Attributes inherited from cldnn::primitive
const primitive_type_id	type
	Primitive's type id.

const primitive_id	id
	Primitive's id.

fixed_size_vector_ref	input
	List of ids of input primitives.

padding	output_padding
	Requested output padding.

Additional Inherited Members
Protected Member Functions inherited from cldnn::primitive_base< prior_box, cldnn_prior_box_desc >
	primitive_base (const primitive_id &id, const std::vector< primitive_id > &input, const padding &output_padding=padding())

	primitive_base (const cldnn_prior_box_desc *dto)

Protected Member Functions inherited from cldnn::primitive
virtual std::vector< std::reference_wrapper< const primitive_id > >	get_dependencies () const

Protected Attributes inherited from cldnn::primitive
primitive_id_arr	_input

Detailed Description

Generates a set of default bounding boxes with different sizes and aspect ratios.

The prior-boxes are shared across all the images in a batch (since they have the same width and height). First feature stores the mean of each prior coordinate. Second feature stores the variance of each prior coordinate.

Definition at line 38 of file prior_box.hpp.

Constructor & Destructor Documentation

◆ prior_box()

cldnn::prior_box::prior_box	(	const primitive_id &	id,
		const primitive_id &	input,
		const tensor &	img_size,
		const std::vector< float > &	min_sizes,
		const std::vector< float > &	max_sizes = `{}`,
		const std::vector< float > &	aspect_ratios = `{}`,
		const bool	flip = `true`,
		const bool	clip = `false`,
		const std::vector< float > &	variance = `{}`,
		const float	step_width = `0.f`,
		const float	step_height = `0.f`,
		const float	offset = `0.5f`,
		const padding &	output_padding = `padding()`
	)

inline

Constructs prior-box primitive.

Parameters

id	This primitive id.
input	Input primitive id.
img_size	Image width and height.
min_sizes	Minimum box sizes in pixels.
max_sizes	Maximum box sizes in pixels.
aspect_ratios	Various of aspect ratios. Duplicate ratios will be ignored.
flip	If true, will flip each aspect ratio. For example, if there is aspect ratio "r", aspect ratio "1.0/r" we will generated as well.
clip	If true, will clip the prior so that it is within [0, 1].
variance	Variance for adjusting the prior boxes.
step_width	Step width.
step_height	Step height.
offset	Offset to the top left corner of each cell.

Definition at line 55 of file prior_box.hpp.

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

C:/ML/OpenSource/clDNN/api/CPP/prior_box.hpp

Public Types

Public Member Functions

Static Public Member Functions

Public Attributes

Additional Inherited Members

Detailed Description

Constructor & Destructor Documentation

◆ prior_box()