Motion

Simd::Motion is C++ framework for motion detection. More...

Namespaces
namespace	Simd::Motion
	Contains a C++ framework for motion detection.

Data Structures
struct	Position
	Position of a detected object at one timestamp. More...
struct	Object
	A classified moving object on the current frame. More...
struct	Event
	An event generated by Simd::Motion::Detector on the current frame. More...
struct	Metadata
	Result of Detector::NextFrame for the current frame. More...
struct	Model
	Scene model used to calibrate Simd::Motion::Detector. More...
struct	Options
	Options used by Simd::Motion::Detector. More...
class	Detector
	Motion detector. More...

Functions
SIMD_INLINE double	ScreenToOnvifX (ptrdiff_t x, ptrdiff_t screenWidth)
	Converts a screen X-coordinate to an ONVIF X-coordinate. More...
SIMD_INLINE double	ScreenToOnvifY (ptrdiff_t y, ptrdiff_t screenHeight)
	Converts a screen Y-coordinate to an ONVIF Y-coordinate. More...
SIMD_INLINE FPoint	ScreenToOnvif (const Point &point, const Point &screenSize)
	Converts screen 2D-coordinates to ONVIF 2D-coordinates. More...
SIMD_INLINE FSize	ScreenToOnvifSize (const Size &size, const Point &screenSize)
	Converts a screen 2D-size to an ONVIF 2D-size. More...
SIMD_INLINE ptrdiff_t	OnvifToScreenX (double x, ptrdiff_t screenWidth)
	Converts an ONVIF X-coordinate to a screen X-coordinate. More...
SIMD_INLINE ptrdiff_t	OnvifToScreenY (double y, ptrdiff_t screenHeight)
	Converts an ONVIF Y-coordinate to a screen Y-coordinate. More...
SIMD_INLINE Point	OnvifToScreen (const FPoint &point, const Point &screenSize)
	Converts ONVIF 2D-coordinates to screen 2D-coordinates. More...
SIMD_INLINE Size	OnvifToScreenSize (const FSize &size, const Point &screenSize)
	Converts an ONVIF 2D-size to a screen 2D-size. More...
SIMD_INLINE String	ToString (Id id)
	Converts an object ID to a string. More...

Detailed Description

Simd::Motion is C++ framework for motion detection.

Function Documentation

ScreenToOnvifX()

SIMD_INLINE double ScreenToOnvifX	(	ptrdiff_t	x,
		ptrdiff_t	screenWidth
	)

Converts a screen X-coordinate to an ONVIF X-coordinate.

Maps pixel x in [0, screenWidth] to [-1, 1]. Left edge is -1, right edge is 1. Typical usage is ScreenToOnvif() when filling Model::roi from pixel points.

Parameters

[in]	x	- a screen X-coordinate in pixels.
[in]	screenWidth	- a screen width in pixels.

Returns

ONVIF X-coordinate in range [-1, 1].

ScreenToOnvifY()

SIMD_INLINE double ScreenToOnvifY	(	ptrdiff_t	y,
		ptrdiff_t	screenHeight
	)

Converts a screen Y-coordinate to an ONVIF Y-coordinate.

Maps pixel y in [0, screenHeight] to [1, -1]. Top edge is 1, bottom edge is -1 (ONVIF Y grows up). Typical usage is ScreenToOnvif() when filling Model::roi.

Parameters

[in]	y	- a screen Y-coordinate in pixels.
[in]	screenHeight	- a screen height in pixels.

Returns

ONVIF Y-coordinate in range [-1, 1].

ScreenToOnvif()

SIMD_INLINE FPoint ScreenToOnvif	(	const Point &	point,
		const Point &	screenSize
	)

Converts screen 2D-coordinates to ONVIF 2D-coordinates.

Combines ScreenToOnvifX and ScreenToOnvifY. Typical usage converts a pixel polygon to Model::roi. Detector uses the inverse OnvifToScreen when it applies the model.

Parameters

[in]	point	- a screen point in pixels.
[in]	screenSize	- a screen size in pixels (width and height).

Returns

ONVIF point in range [-1, 1].

ScreenToOnvifSize()

SIMD_INLINE FSize ScreenToOnvifSize	(	const Size &	size,
		const Point &	screenSize
	)

Converts a screen 2D-size to an ONVIF 2D-size.

Full screen maps to (2, 2). Typical usage converts a pixel object size to Model::size. Detector uses the inverse OnvifToScreenSize for Model::size.

Parameters

[in]	size	- a screen size in pixels (width and height).
[in]	screenSize	- a screen size in pixels (width and height).

Returns

ONVIF size in range [0, 2].

OnvifToScreenX()

SIMD_INLINE ptrdiff_t OnvifToScreenX	(	double	x,
		ptrdiff_t	screenWidth
	)

Converts an ONVIF X-coordinate to a screen X-coordinate.

Maps x in [-1, 1] to pixels and clamps the result to [0, screenWidth - 1]. Detector uses this through OnvifToScreen when it applies Model::roi.

Parameters

[in]	x	- an ONVIF X-coordinate in range [-1, 1].
[in]	screenWidth	- a screen width in pixels.

Returns

screen X-coordinate in pixels.

OnvifToScreenY()

SIMD_INLINE ptrdiff_t OnvifToScreenY	(	double	y,
		ptrdiff_t	screenHeight
	)

Converts an ONVIF Y-coordinate to a screen Y-coordinate.

Maps y in [-1, 1] to pixels (ONVIF +Y is up) and clamps the result to [0, screenHeight - 1]. Detector uses this through OnvifToScreen when it applies Model::roi.

Parameters

[in]	y	- an ONVIF Y-coordinate in range [-1, 1].
[in]	screenHeight	- a screen height in pixels.

Returns

screen Y-coordinate in pixels.

OnvifToScreen()

SIMD_INLINE Point OnvifToScreen	(	const FPoint &	point,
		const Point &	screenSize
	)

Converts ONVIF 2D-coordinates to screen 2D-coordinates.

Combines OnvifToScreenX and OnvifToScreenY. Detector uses this to convert Model::roi vertices to pixels.

Parameters

[in]	point	- an ONVIF point in range [-1, 1].
[in]	screenSize	- a screen size in pixels (width and height).

Returns

screen point in pixels.

OnvifToScreenSize()

SIMD_INLINE Size OnvifToScreenSize	(	const FSize &	size,
		const Point &	screenSize
	)

Converts an ONVIF 2D-size to a screen 2D-size.

Detector uses this to convert Model::size to a pixel object size that sets the minimum motion-region area. Debug annotation of the model draws that minimum rectangle.

Parameters

[in]	size	- an ONVIF size in range [0, 2].
[in]	screenSize	- a screen size in pixels (width and height).

Returns

screen size in pixels.

ToString()

SIMD_INLINE String ToString

(

Id

id

)

Converts an object ID to a string.

Typical usage draws ToString(object.id) above Object::rect with Simd::Font.

Parameters

[in]

id

- an object ID (Object::id or Event::objectId).

Returns

decimal string representation of the ID.