DlibFaceLandmarkDetector.FaceLandmarkDetector Class Reference

Face landmark detector. More...

Classes
class	RectDetection
	Represents the detection result of an object, including its bounding rectangle, detection confidence, and weight index. More...

Public Member Functions
	FaceLandmarkDetector (string objectDetectorFilePath, string shapePredictorFilePath)
	Initializes a new instance of the FaceLandmarkDetector class.
	FaceLandmarkDetector (string shapePredictorFilePath)
	Initializes a new instance of the FaceLandmarkDetector class.
List< Rect >	Detect ()
	Detects objects in the image.
List< Rect >	Detect (double adjust_threshold)
	Detects objects in the image.
double[]	DetectArray ()
	Detects objects in the image.
double[]	DetectArray (double adjust_threshold)
	Detects objects in the image.
List< Vector2 >	DetectLandmark (double left, double top, double width, double height)
	Detects the object landmarks within the specified region.
double double y[]	DetectLandmark (in(double x, double y, double width, double height) rect)
List< Vector2 >	DetectLandmark (Rect rect)
	Detects the object landmarks within the specified region defined by a rectangle.
double[]	DetectLandmarkArray (double left, double top, double width, double height)
	Detects the object landmarks within the specified region defined by a rectangle.
double[]	DetectLandmarkArray (in(double x, double y, double width, double height) rect)
	Detects the object landmarks within the specified region defined by a rectangle.
double[]	DetectLandmarkArray (Rect rect)
	Detects the object landmarks within the specified region defined by a rectangle.
int	DetectLandmarkOnly (double left, double top, double width, double height)
	Detects objects and returns the number of detected object landmarks.
int	DetectLandmarkOnly (in(double x, double y, double width, double height) rect)
	Detects objects and returns the number of detected object landmarks within the specified region defined by a rectangle.
int	DetectLandmarkOnly (Rect rect)
	Detects objects and returns the number of detected object landmarks within the specified region defined by a rectangle.
int	DetectOnly ()
	Detects objects and returns the number of objects detected.
int	DetectOnly (double adjust_threshold)
	Detects objects and returns the number of objects detected.
List< RectDetection >	DetectRectDetection ()
	Detects objects in the image.
List< RectDetection >	DetectRectDetection (double adjust_threshold)
	Detects objects in the image.
double double double double height[]	DetectValueTuple ()
double double double double height[]	DetectValueTuple (double adjust_threshold)
void	DrawDetectLandmarkResult (IntPtr intPtr, int width, int height, int bytesPerPixel, bool flip, int r, int g, int b, int a)
	Draws the detected landmark result on the provided image buffer with optional vertical flipping.
void	DrawDetectLandmarkResult (IntPtr intPtr, int width, int height, int bytesPerPixel, int r, int g, int b, int a)
	Draws the detected landmark result on the provided image buffer.
void	DrawDetectLandmarkResult (Texture2D texture2D, int r, int g, int b, int a, bool updateMipmaps=false, bool makeNoLongerReadable=false)
	Draws the detected landmark result on the provided texture.
void	DrawDetectLandmarkResult (Texture2D texture2D, int r, int g, int b, int a, Color32[] pixels32Buffer, byte[] rawTextureDataBuffer=null, bool updateMipmaps=false, bool makeNoLongerReadable=false)
	Draws the detected landmark result on the provided texture.
void	DrawDetectLandmarkResult< T > (T[] array, int width, int height, int bytesPerPixel, bool flip, int r, int g, int b, int a)
	Draws the detected landmark result on the provided image buffer, with an option to flip the image vertically.
void	DrawDetectLandmarkResult< T > (T[] array, int width, int height, int bytesPerPixel, int r, int g, int b, int a)
	Draws the detected landmark result on the provided image buffer.
void	DrawDetectResult (IntPtr intPtr, int width, int height, int bytesPerPixel, bool flip, int r, int g, int b, int a, int thickness)
	Draws the detection result on a memory buffer with optional vertical flipping.
void	DrawDetectResult (IntPtr intPtr, int width, int height, int bytesPerPixel, int r, int g, int b, int a, int thickness)
	Draws the detection result on a memory buffer.
void	DrawDetectResult (Texture2D texture2D, int r, int g, int b, int a, int thickness, bool updateMipmaps=false, bool makeNoLongerReadable=false)
	Draws the detection result on the specified texture.
void	DrawDetectResult (Texture2D texture2D, int r, int g, int b, int a, int thickness, Color32[] pixels32Buffer, byte[] rawTextureDataBuffer=null, bool updateMipmaps=false, bool makeNoLongerReadable=false)
	Draws the detection result on the specified texture.
void	DrawDetectResult< T > (T[] array, int width, int height, int bytesPerPixel, bool flip, int r, int g, int b, int a, int thickness)
	Draws the detection result on an array representing the image data.
void	DrawDetectResult< T > (T[] array, int width, int height, int bytesPerPixel, int r, int g, int b, int a, int thickness)
	Draws the detection result on an array representing the image data.
void	GetDetectLandmarkResult (double[] result)
	Gets the result data of the object landmarks detected by the DetectLandmarkOnly() method, passing a data size of DetectLandmarkOnly() * 2 as an argument. This method can retrieve results without memory allocation.
void	GetDetectLandmarkResult (Span< double > result)
	Gets the result data of the object landmarks detected by the DetectLandmarkOnly() method, passing a data size of DetectLandmarkOnly() * 2 as an argument. This method can retrieve results without memory allocation.
void	GetDetectResult (double[] result)
	Gets the result data of the objects detected by the DetectOnly() method, passing a data size of DetectOnly() * 6 as an argument. This method can retrieve results without memory allocation.
void	GetDetectResult (Span< double > result)
	Gets the result data of the objects detected by the DetectOnly() method, passing a data size of DetectOnly() * 6 as an argument. This method can retrieve results without memory allocation.
long	GetShapePredictorNumFeatures ()
	Gets the number of features in the shape predictor.
long	GetShapePredictorNumParts ()
	Gets the number of parts in the shape predictor.
bool	IsAllPartsInRect ()
	Determines whether all of the object parts are contained within the object rectangle.
void	SetImage (IntPtr intPtr, int width, int height, int bytesPerPixel)
	Sets the image from an IntPtr.
void	SetImage (IntPtr intPtr, int width, int height, int bytesPerPixel, bool flip)
	Sets the image from an IntPtr.
void	SetImage (Texture2D texture2D)
	Sets the image from a Texture2D.
void	SetImage (WebCamTexture webCamTexture)
	Sets the image from a WebCamTexture.
void	SetImage (WebCamTexture webCamTexture, Color32[] pixels32Buffer)
	Sets the image from a WebCamTexture.
void	SetImage< T > (T[] array, int width, int height, int bytesPerPixel)
	Sets the image from a pixel data array.
void	SetImage< T > (T[] array, int width, int height, int bytesPerPixel, bool flip)
	Sets the image from a pixel data array.
Public Member Functions inherited from DlibFaceLandmarkDetector.DisposableObject
void	Dispose ()
void	ThrowIfDisposed ()

Public Attributes
double double double	width
double	x
	Detects objects in the image.
double double	y

Protected Member Functions
override void	Dispose (bool disposing)
Protected Member Functions inherited from DlibFaceLandmarkDetector.DisposableDlibObject
	DisposableDlibObject ()
	DisposableDlibObject (bool isEnabledDispose)
	DisposableDlibObject (IntPtr ptr)
	DisposableDlibObject (IntPtr ptr, bool isEnabledDispose)
Protected Member Functions inherited from DlibFaceLandmarkDetector.DisposableObject
	DisposableObject ()
	DisposableObject (bool isEnabledDispose)

Additional Inherited Members
Static Public Member Functions inherited from DlibFaceLandmarkDetector.DisposableObject
static IntPtr	ThrowIfNullIntPtr (IntPtr ptr)
Package Attributes inherited from DlibFaceLandmarkDetector.DisposableDlibObject
Properties inherited from DlibFaceLandmarkDetector.DisposableObject
bool	IsDisposed [get, protected set]
bool	IsEnabledDispose [get, set]

Detailed Description

Face landmark detector.

Constructor & Destructor Documentation

FaceLandmarkDetector() [1/2]

DlibFaceLandmarkDetector.FaceLandmarkDetector.FaceLandmarkDetector

(

string

shapePredictorFilePath

)

Initializes a new instance of the FaceLandmarkDetector class.

This instance uses the default frontal face detector.
ObjectDetector is initialized as follows:

frontal_face_detector face_detector;
face_detector = get_frontal_face_detector();

ShapePredictor is initialized as follows:

shape_predictor sp;
deserialize(shape_predictor_filename) >> sp;

Parameters

shapePredictorFilePath

The file path of the shape predictor.

FaceLandmarkDetector() [2/2]

DlibFaceLandmarkDetector.FaceLandmarkDetector.FaceLandmarkDetector	(	string	objectDetectorFilePath,
		string	shapePredictorFilePath )

Initializes a new instance of the FaceLandmarkDetector class.

ObjectDetector is initialized with the following code:

if (object_detector_filename != null)
{
    object_detector<scan_fhog_pyramid<pyramid_down<6>>> simple_detector;
    deserialize(object_detector_filename) >> simple_detector;
}
else
{
    frontal_face_detector face_detector;
    face_detector = get_frontal_face_detector();
}

ShapePredictor is initialized with the following code:

shape_predictor sp;
deserialize(shape_predictor_filename) >> sp;

Parameters

objectDetectorFilePath	The file path of the object detector.
shapePredictorFilePath	The file path of the shape predictor.

Member Function Documentation

Detect() [1/2]

List< Rect > DlibFaceLandmarkDetector.FaceLandmarkDetector.Detect

(

)

Detects objects in the image.

Returns

A list of Rect representing the detected objects.

Detect() [2/2]

List< Rect > DlibFaceLandmarkDetector.FaceLandmarkDetector.Detect

(

double

adjust_threshold

)

Detects objects in the image.

Parameters

adjust_threshold

A parameter to adjust the detection threshold.

Returns

A list of Rect representing the detected objects.

DetectArray() [1/2]

double[] DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectArray

(

)

Detects objects in the image.

Returns

An array of doubles representing detected object data. The array contains values in the following order:

• left_0, top_0, width_0, height_0, detection_confidence_0, weight_index_0
• left_1, top_1, width_1, height_1, detection_confidence_1, weight_index_1
• ...

DetectArray() [2/2]

double[] DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectArray

(

double

adjust_threshold

)

Detects objects in the image.

Parameters

adjust_threshold

A parameter to adjust the detection threshold.

Returns

An array of doubles representing detected object data. The array contains values in the following order:

• left_0, top_0, width_0, height_0, detection_confidence_0, weight_index_0
• left_1, top_1, width_1, height_1, detection_confidence_1, weight_index_1
• ...

DetectLandmark() [1/3]

List< Vector2 > DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectLandmark	(	double	left,
		double	top,
		double	width,
		double	height )

Detects the object landmarks within the specified region.

Parameters

left	The left coordinate of the region to search for landmarks.
top	The top coordinate of the region to search for landmarks.
width	The width of the region to search for landmarks.
height	The height of the region to search for landmarks.

Returns

A list of Vector2 representing the detected object landmarks.

DetectLandmark() [2/3]

double double y[] DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectLandmark

(

in(double x, double y, double width, double height)

rect

)

DetectLandmark() [3/3]

List< Vector2 > DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectLandmark

(

Rect

rect

)

Detects the object landmarks within the specified region defined by a rectangle.

Parameters

rect	A rectangle defining the region to search for landmarks. The rectangle is defined by its left, top, width, and height.

Returns

A list of Vector2 representing the detected object landmarks.

DetectLandmarkArray() [1/3]

double[] DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectLandmarkArray	(	double	left,
		double	top,
		double	width,
		double	height )

Detects the object landmarks within the specified region defined by a rectangle.

Parameters

left	The left coordinate of the rectangle.
top	The top coordinate of the rectangle.
width	The width of the rectangle.
height	The height of the rectangle.

Returns

An array of doubles representing the detected object landmark data. The array contains values in the following order:

• x_0, y_0
• x_1, y_1
• ...

DetectLandmarkArray() [2/3]

double[] DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectLandmarkArray

(

in(double x, double y, double width, double height)

rect

)

Detects the object landmarks within the specified region defined by a rectangle.

Parameters

rect	A ValueTuple<Double, Double, Double, Double> representing a rectangle. The ValueTuple contains the following values: x (left), y (top), width, height

Returns

An array of doubles representing the detected object landmark data. The array contains values in the following order:

• x_0, y_0
• x_1, y_1
• ...

If no landmarks are detected, the method returns the default value (null).

DetectLandmarkArray() [3/3]

double[] DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectLandmarkArray

(

Rect

rect

)

Detects the object landmarks within the specified region defined by a rectangle.

Parameters

rect	The rectangle defining the region in which to detect object landmarks.

Returns

An array of doubles representing the detected object landmark data. The array contains values in the following order:

• x_0, y_0
• x_1, y_1
• ...

DetectLandmarkOnly() [1/3]

int DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectLandmarkOnly	(	double	left,
		double	top,
		double	width,
		double	height )

Detects objects and returns the number of detected object landmarks.

Parameters

left	The left coordinate of the object region to detect landmarks.
top	The top coordinate of the object region to detect landmarks.
width	The width of the object region to detect landmarks.
height	The height of the object region to detect landmarks.

Returns

An integer representing the number of detected object landmarks.

DetectLandmarkOnly() [2/3]

int DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectLandmarkOnly

(

in(double x, double y, double width, double height)

rect

)

Detects objects and returns the number of detected object landmarks within the specified region defined by a rectangle.

Parameters

rect	A ValueTuple<Double, Double, Double, Double> representing a rectangle. The ValueTuple contains the following values: x (left), y (top), width, height

Returns

An integer representing the number of object landmarks detected within the specified region.

DetectLandmarkOnly() [3/3]

int DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectLandmarkOnly

(

Rect

rect

)

Detects objects and returns the number of detected object landmarks within the specified region defined by a rectangle.

Parameters

rect	The rectangle defining the region in which to detect object landmarks.

Returns

An integer representing the number of detected object landmarks.

DetectOnly() [1/2]

int DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectOnly

(

)

Detects objects and returns the number of objects detected.

Returns

The number of objects detected.

DetectOnly() [2/2]

int DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectOnly

(

double

adjust_threshold

)

Detects objects and returns the number of objects detected.

Parameters

adjust_threshold

A parameter to adjust the detection threshold.

Returns

The number of objects detected.

DetectRectDetection() [1/2]

List< RectDetection > DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectRectDetection

(

)

Detects objects in the image.

Returns

A list of RectDetection representing the detected objects.

DetectRectDetection() [2/2]

List< RectDetection > DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectRectDetection

(

double

adjust_threshold

)

Detects objects in the image.

Parameters

adjust_threshold

A parameter to adjust the detection threshold.

Returns

A list of RectDetection representing the detected objects.

DetectValueTuple() [1/2]

double double double double height[] DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectValueTuple

(

)

DetectValueTuple() [2/2]

double double double double height[] DlibFaceLandmarkDetector.FaceLandmarkDetector.DetectValueTuple

(

double

adjust_threshold

)

Dispose()

override void DlibFaceLandmarkDetector.FaceLandmarkDetector.Dispose ( bool disposing )

protectedvirtual

Reimplemented from DlibFaceLandmarkDetector.DisposableDlibObject.

DrawDetectLandmarkResult() [1/4]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.DrawDetectLandmarkResult	(	IntPtr	intPtr,
		int	width,
		int	height,
		int	bytesPerPixel,
		bool	flip,
		int	r,
		int	g,
		int	b,
		int	a )

Draws the detected landmark result on the provided image buffer with optional vertical flipping.

Parameters

intPtr	The pointer to the memory buffer where the detected landmark result will be drawn.
width	The width of the image (in pixels) where the landmarks will be drawn.
height	The height of the image (in pixels) where the landmarks will be drawn.
bytesPerPixel	The number of bytes per pixel in the memory buffer. Valid values are 1 (grayscale), 3 (RGB), or 4 (RGBA).
flip	A boolean flag indicating whether to flip the image vertically. Set to true to flip, or false to leave the image unflipped.
r	The red component (0-255) of the color to use for drawing the landmarks.
g	The green component (0-255) of the color to use for drawing the landmarks.
b	The blue component (0-255) of the color to use for drawing the landmarks.
a	The alpha (transparency) component (0-255) of the color to use for drawing the landmarks.

DrawDetectLandmarkResult() [2/4]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.DrawDetectLandmarkResult	(	IntPtr	intPtr,
		int	width,
		int	height,
		int	bytesPerPixel,
		int	r,
		int	g,
		int	b,
		int	a )

Draws the detected landmark result on the provided image buffer.

Parameters

intPtr	The pointer to the memory buffer where the detected landmark result will be drawn.
width	The width of the image (in pixels) where the landmarks will be drawn.
height	The height of the image (in pixels) where the landmarks will be drawn.
bytesPerPixel	The number of bytes per pixel in the memory buffer. Valid values are 1 (grayscale), 3 (RGB), or 4 (RGBA).
r	The red component (0-255) of the color to use for drawing the landmarks.
g	The green component (0-255) of the color to use for drawing the landmarks.
b	The blue component (0-255) of the color to use for drawing the landmarks.
a	The alpha (transparency) component (0-255) of the color to use for drawing the landmarks.

DrawDetectLandmarkResult() [3/4]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.DrawDetectLandmarkResult	(	Texture2D	texture2D,
		int	r,
		int	g,
		int	b,
		int	a,
		bool	updateMipmaps = false,
		bool	makeNoLongerReadable = false )

Draws the detected landmark result on the provided texture.

Parameters

texture2D	The Texture2D on where the detected landmark result will be drawn. Processing speed is fastest when the texture format is TextureFormat.RGBA32, TextureFormat.RGB24, or TextureFormat.Alpha8.
r	The red component (0-255) of the color to use for drawing the landmarks.
g	The green component (0-255) of the color to use for drawing the landmarks.
b	The blue component (0-255) of the color to use for drawing the landmarks.
a	The alpha (transparency) component (0-255) of the color to use for drawing the landmarks.
updateMipmaps	If set to true, mipmap levels of the texture will be recalculated.
makeNoLongerReadable	If set to true, the system memory copy of the texture will be released, making it no longer readable.

DrawDetectLandmarkResult() [4/4]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.DrawDetectLandmarkResult	(	Texture2D	texture2D,
		int	r,
		int	g,
		int	b,
		int	a,
		Color32[]	pixels32Buffer,
		byte[]	rawTextureDataBuffer = null,
		bool	updateMipmaps = false,
		bool	makeNoLongerReadable = false )

Draws the detected landmark result on the provided texture.

Parameters

texture2D	The Texture2D on where the detected landmark result will be drawn. Processing speed is fastest when the texture format is TextureFormat.RGBA32, TextureFormat.RGB24, or TextureFormat.Alpha8.
r	The red component (0-255) of the color to use for drawing the landmarks.
g	The green component (0-255) of the color to use for drawing the landmarks.
b	The blue component (0-255) of the color to use for drawing the landmarks.
a	The alpha (transparency) component (0-255) of the color to use for drawing the landmarks.
pixels32Buffer	An optional array to receive pixel data in the form of Color32. You can pass in an existing array to avoid allocating new memory each frame. The array should be initialized to a length matching the width * height of the texture. WebCamTexture.GetPixels32
rawTextureDataBuffer	An optional array to receive raw texture data in the form of bytes. You can pass in an existing array to avoid allocating new memory each frame. The array should be initialized to a length matching the raw data size of the texture. Texture2D.GetRawTextureData
updateMipmaps	If set to true, mipmap levels of the texture will be recalculated.
makeNoLongerReadable	If set to true, the system memory copy of the texture will be released, making it no longer readable.

DrawDetectLandmarkResult< T >() [1/2]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.DrawDetectLandmarkResult< T >	(	T[]	array,
		int	width,
		int	height,
		int	bytesPerPixel,
		bool	flip,
		int	r,
		int	g,
		int	b,
		int	a )

Draws the detected landmark result on the provided image buffer, with an option to flip the image vertically.

Parameters

array	The array representing the image data where the detected landmark result will be drawn. The array should contain pixel data in the format corresponding to bytesPerPixel .
width	The width of the image (in pixels) where the landmarks will be drawn.
height	The height of the image (in pixels) where the landmarks will be drawn.
bytesPerPixel	The number of bytes per pixel in the memory buffer. Valid values are 1 (grayscale), 3 (RGB), or 4 (RGBA).
flip	A boolean flag indicating whether to flip the image vertically. Set to true to flip, or false to leave the image unflipped.
r	The red component (0-255) of the color to use for drawing the landmarks.
g	The green component (0-255) of the color to use for drawing the landmarks.
b	The blue component (0-255) of the color to use for drawing the landmarks.
a	The alpha (transparency) component (0-255) of the color to use for drawing the landmarks.

Type Constraints

T

:

unmanaged

DrawDetectLandmarkResult< T >() [2/2]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.DrawDetectLandmarkResult< T >	(	T[]	array,
		int	width,
		int	height,
		int	bytesPerPixel,
		int	r,
		int	g,
		int	b,
		int	a )

Draws the detected landmark result on the provided image buffer.

Parameters

array	The array representing the image data where the detected landmark result will be drawn. The array should contain pixel data in the format corresponding to bytesPerPixel .
width	The width of the image (in pixels) where the landmarks will be drawn.
height	The height of the image (in pixels) where the landmarks will be drawn.
bytesPerPixel	The number of bytes per pixel in the memory buffer. Valid values are 1 (grayscale), 3 (RGB), or 4 (RGBA).
r	The red component (0-255) of the color to use for drawing the landmarks.
g	The green component (0-255) of the color to use for drawing the landmarks.
b	The blue component (0-255) of the color to use for drawing the landmarks.
a	The alpha (transparency) component (0-255) of the color to use for drawing the landmarks.

Type Constraints

T

:

unmanaged

DrawDetectResult() [1/4]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.DrawDetectResult	(	IntPtr	intPtr,
		int	width,
		int	height,
		int	bytesPerPixel,
		bool	flip,
		int	r,
		int	g,
		int	b,
		int	a,
		int	thickness )

Draws the detection result on a memory buffer with optional vertical flipping.

Parameters

intPtr	The pointer to the memory buffer where the detection result will be drawn.
width	The width of the image (in pixels) where the detection result will be drawn.
height	The height of the image (in pixels) where the detection result will be drawn.
bytesPerPixel	The number of bytes per pixel in the memory buffer. Valid values are 1 (grayscale), 3 (RGB), or 4 (RGBA).
flip	A boolean flag indicating whether to flip the image vertically. Set to true to flip, or false to leave the image unflipped.
r	The red component of the color to be used for drawing (0-255).
g	The green component of the color to be used for drawing (0-255).
b	The blue component of the color to be used for drawing (0-255).
a	The alpha (transparency) component of the color to be used for drawing (0-255).
thickness	The thickness of the lines used to draw the detection results.

DrawDetectResult() [2/4]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.DrawDetectResult	(	IntPtr	intPtr,
		int	width,
		int	height,
		int	bytesPerPixel,
		int	r,
		int	g,
		int	b,
		int	a,
		int	thickness )

Draws the detection result on a memory buffer.

Parameters

intPtr	The pointer to the memory buffer where the detection result will be drawn.
width	The width of the image (in pixels) where the detection result will be drawn.
height	The height of the image (in pixels) where the detection result will be drawn.
bytesPerPixel	The number of bytes per pixel in the memory buffer. Valid values are 1 (grayscale), 3 (RGB), or 4 (RGBA).
r	The red component of the color to be used for drawing (0-255).
g	The green component of the color to be used for drawing (0-255).
b	The blue component of the color to be used for drawing (0-255).
a	The alpha (transparency) component of the color to be used for drawing (0-255).
thickness	The thickness of the lines used to draw the detection results.

DrawDetectResult() [3/4]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.DrawDetectResult	(	Texture2D	texture2D,
		int	r,
		int	g,
		int	b,
		int	a,
		int	thickness,
		bool	updateMipmaps = false,
		bool	makeNoLongerReadable = false )

Draws the detection result on the specified texture.

Parameters

texture2D	The Texture2D on which the detection result will be drawn. Processing speed is fastest when the texture format is TextureFormat.RGBA32, TextureFormat.RGB24, or TextureFormat.Alpha8.
r	The red component of the color to be used for drawing (0-255).
g	The green component of the color to be used for drawing (0-255).
b	The blue component of the color to be used for drawing (0-255).
a	The alpha (transparency) component of the color to be used for drawing (0-255).
thickness	The thickness of the lines used to draw the detection results.
updateMipmaps	If set to true, mipmap levels of the texture will be recalculated.
makeNoLongerReadable	If set to true, the system memory copy of the texture will be released, making it no longer readable.

DrawDetectResult() [4/4]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.DrawDetectResult	(	Texture2D	texture2D,
		int	r,
		int	g,
		int	b,
		int	a,
		int	thickness,
		Color32[]	pixels32Buffer,
		byte[]	rawTextureDataBuffer = null,
		bool	updateMipmaps = false,
		bool	makeNoLongerReadable = false )

Draws the detection result on the specified texture.

Parameters

texture2D	The Texture2D on which the detection result will be drawn. Processing speed is fastest when the texture format is TextureFormat.RGBA32, TextureFormat.RGB24, or TextureFormat.Alpha8.
r	The red component of the color to be used for drawing (0-255).
g	The green component of the color to be used for drawing (0-255).
b	The blue component of the color to be used for drawing (0-255).
a	The alpha (transparency) component of the color to be used for drawing (0-255).
thickness	The thickness of the lines used to draw the detection results.
pixels32Buffer	An optional array to receive the Color32 pixel data. You can pass in an array to avoid allocating new memory each frame. The array must be initialized to a length matching the width * height of the texture. WebCamTexture.GetPixels32.
rawTextureDataBuffer	An optional array to receive raw texture data. You can pass in an array to avoid allocating new memory each frame. The array must be initialized to a length matching the raw data size of the texture. Texture2D.GetRawTextureData.
updateMipmaps	If set to true, mipmap levels of the texture will be recalculated.
makeNoLongerReadable	If set to true, the system memory copy of the texture will be released, making it no longer readable.

DrawDetectResult< T >() [1/2]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.DrawDetectResult< T >	(	T[]	array,
		int	width,
		int	height,
		int	bytesPerPixel,
		bool	flip,
		int	r,
		int	g,
		int	b,
		int	a,
		int	thickness )

Draws the detection result on an array representing the image data.

Parameters

array	The array representing the image data where the detection result will be drawn. The array should contain pixel data in the format corresponding to bytesPerPixel .
width	The width of the image (in pixels) where the detection result will be drawn.
height	The height of the image (in pixels) where the detection result will be drawn.
bytesPerPixel	The number of bytes per pixel in the memory buffer. Valid values are 1 (grayscale), 3 (RGB), or 4 (RGBA).
flip	A boolean flag indicating whether to flip the image vertically. Set to true to flip, or false to leave the image unflipped.
r	The red component of the color to be used for drawing (0-255).
g	The green component of the color to be used for drawing (0-255).
b	The blue component of the color to be used for drawing (0-255).
a	The alpha (transparency) component of the color to be used for drawing (0-255).
thickness	The thickness of the lines used to draw the detection results.

Type Constraints

T

:

unmanaged

DrawDetectResult< T >() [2/2]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.DrawDetectResult< T >	(	T[]	array,
		int	width,
		int	height,
		int	bytesPerPixel,
		int	r,
		int	g,
		int	b,
		int	a,
		int	thickness )

Draws the detection result on an array representing the image data.

Parameters

array	The array representing the image data where the detection result will be drawn. The array should contain pixel data in the format corresponding to bytesPerPixel .
width	The width of the image (in pixels) where the detection result will be drawn.
height	The height of the image (in pixels) where the detection result will be drawn.
bytesPerPixel	The number of bytes per pixel in the memory buffer. Valid values are 1 (grayscale), 3 (RGB), or 4 (RGBA).
r	The red component of the color to be used for drawing (0-255).
g	The green component of the color to be used for drawing (0-255).
b	The blue component of the color to be used for drawing (0-255).
a	The alpha (transparency) component of the color to be used for drawing (0-255).
thickness	The thickness of the lines used to draw the detection results.

Type Constraints

T

:

unmanaged

GetDetectLandmarkResult() [1/2]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.GetDetectLandmarkResult

(

double[]

result

)

Gets the result data of the object landmarks detected by the DetectLandmarkOnly() method, passing a data size of DetectLandmarkOnly() * 2 as an argument. This method can retrieve results without memory allocation.

Parameters

result

An array of doubles representing detected object landmark data. The array contains values in the following order: x_0, y_0 x_1, y_1 ... The array is passed by reference to receive the landmark result data without allocating new memory each time.

GetDetectLandmarkResult() [2/2]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.GetDetectLandmarkResult

(

Span< double >

result

)

Gets the result data of the object landmarks detected by the DetectLandmarkOnly() method, passing a data size of DetectLandmarkOnly() * 2 as an argument. This method can retrieve results without memory allocation.

Parameters

result

A span of doubles representing detected object landmark data. The span contains values in the following order: x_0, y_0 x_1, y_1 ... The span is passed by reference to receive the landmark result data without allocating new memory each time.

GetDetectResult() [1/2]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.GetDetectResult

(

double[]

result

)

Gets the result data of the objects detected by the DetectOnly() method, passing a data size of DetectOnly() * 6 as an argument. This method can retrieve results without memory allocation.

Parameters

result

An array of doubles representing detected object data. The array contains values in the following order: left_0, top_0, width_0, height_0, detection_confidence_0, weight_index_0 left_1, top_1, width_1, height_1, detection_confidence_1, weight_index_1 ... The array is passed by reference to receive the detection result data without allocating new memory each time.

GetDetectResult() [2/2]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.GetDetectResult

(

Span< double >

result

)

Gets the result data of the objects detected by the DetectOnly() method, passing a data size of DetectOnly() * 6 as an argument. This method can retrieve results without memory allocation.

Parameters

result

A span of doubles representing detected object data. The span contains values in the following order: left_0, top_0, width_0, height_0, detection_confidence_0, weight_index_0 left_1, top_1, width_1, height_1, detection_confidence_1, weight_index_1 ... The span is passed by reference to receive the detection result data without allocating new memory each time.

GetShapePredictorNumFeatures()

long DlibFaceLandmarkDetector.FaceLandmarkDetector.GetShapePredictorNumFeatures

(

)

Gets the number of features in the shape predictor.

Returns

A long representing the number of features in the shape predictor.

GetShapePredictorNumParts()

long DlibFaceLandmarkDetector.FaceLandmarkDetector.GetShapePredictorNumParts

(

)

Gets the number of parts in the shape predictor.

Returns

A long representing the number of parts in the shape predictor.

IsAllPartsInRect()

bool DlibFaceLandmarkDetector.FaceLandmarkDetector.IsAllPartsInRect

(

)

Determines whether all of the object parts are contained within the object rectangle.

Returns

A boolean value indicating whether all of the object parts are inside the rectangle. true if all points are contained within the rectangle, otherwise false.

SetImage() [1/5]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.SetImage	(	IntPtr	intPtr,
		int	width,
		int	height,
		int	bytesPerPixel )

Sets the image from an IntPtr.

Parameters

intPtr	The IntPtr pointing to the image data.
width	The width of the image.
height	The height of the image.
bytesPerPixel	The number of bytes per pixel (1, 3, or 4).

SetImage() [2/5]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.SetImage	(	IntPtr	intPtr,
		int	width,
		int	height,
		int	bytesPerPixel,
		bool	flip )

Sets the image from an IntPtr.

Parameters

intPtr	The IntPtr pointing to the image data.
width	The width of the image.
height	The height of the image.
bytesPerPixel	The number of bytes per pixel (1, 3, or 4).
flip	If true, the image will be flipped vertically.

SetImage() [3/5]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.SetImage

(

Texture2D

texture2D

)

Sets the image from a Texture2D.

Parameters

texture2D

Processing speed is fastest when the TextureFormat is RGBA32, RGB24, or Alpha8.

SetImage() [4/5]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.SetImage

(

WebCamTexture

webCamTexture

)

Sets the image from a WebCamTexture.

Parameters

webCamTexture

The WebCamTexture to set as the image.

SetImage() [5/5]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.SetImage	(	WebCamTexture	webCamTexture,
		Color32[]	pixels32Buffer )

Sets the image from a WebCamTexture.

Parameters

webCamTexture	The WebCamTexture to set as the image.
pixels32Buffer	The optional array to receive pixel data. You can pass in an array of Color32 values to avoid allocating new memory each frame. The array should have a length matching the width * height of the texture. WebCamTexture.GetPixels32

SetImage< T >() [1/2]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.SetImage< T >	(	T[]	array,
		int	width,
		int	height,
		int	bytesPerPixel )

Sets the image from a pixel data array.

Parameters

array	The array containing pixel data.
width	The width of the image.
height	The height of the image.
bytesPerPixel	The number of bytes per pixel (1, 3, or 4).

Type Constraints

T

:

unmanaged

SetImage< T >() [2/2]

void DlibFaceLandmarkDetector.FaceLandmarkDetector.SetImage< T >	(	T[]	array,
		int	width,
		int	height,
		int	bytesPerPixel,
		bool	flip )

Sets the image from a pixel data array.

Parameters

array	The array containing pixel data.
width	The width of the image.
height	The height of the image.
bytesPerPixel	The number of bytes per pixel (1, 3, or 4).
flip	If true, the image will be flipped vertically.

Type Constraints

T

:

unmanaged

Member Data Documentation

width

double double double DlibFaceLandmarkDetector.FaceLandmarkDetector.width

x

double DlibFaceLandmarkDetector.FaceLandmarkDetector.x

Detects objects in the image.

Detects the object landmarks within the specified region defined by a rectangle.

Returns

An array of ValueTuple<Double, Double, Double, Double> representing detected object data. Each ValueTuple contains values in the following order:

• x_0 (left), y_0 (top), width_0, height_0
• x_1 (left), y_1 (top), width_1, height_1
• ...

Parameters

adjust_threshold

A parameter to adjust the detection threshold.

Returns

An array of ValueTuple<Double, Double, Double, Double> representing detected object data. Each ValueTuple contains values in the following order:

• x_0 (left), y_0 (top), width_0, height_0
• x_1 (left), y_1 (top), width_1, height_1
• ...

Parameters

rect	A ValueTuple<Double, Double, Double, Double> representing a rectangle. The ValueTuple contains the following values: x (left), y (top), width, height

Returns

An array of ValueTuple<Double, Double> representing the detected object landmarks. Each ValueTuple contains the x and y coordinates of a detected landmark. If no landmarks are detected, the method returns the default value (null).

y

double double DlibFaceLandmarkDetector.FaceLandmarkDetector.y

---

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

• DlibFaceLandmarkDetector/Assets/DlibFaceLandmarkDetector/Scripts/FaceLandmarkDetector.cs