It holds all functions and data related to Face manipulation. More...

#include <Face.h>

Collaboration diagram for Face:

Classes
class	Keypoint
	It holds all attributes related to a facial key point. More...
class	Keypoints
	It holds all facial keypoints detected on a FaceCapture image. More...

Public Types
enum class	AgeEstimationMode : uint8_t { BALANCED = 1 , ACCURATE = 2 }
	Specifies the level of compromise between age estimation speed and accuracy. More...
enum class	GenderEstimationMode : uint8_t { BALANCED = 1 , ACCURATE = 2 }
	Specifies the level of compromise between age estimation speed and accuracy. More...
enum class	PassiveLivenessMode : uint8_t { FAST , ACCURATE }
	Mode of passive liveness calculation. More...
enum class	AttributeID : uint8_t { CONFIDENCE , SHARPNESS , BRIGHTNESS , CONTRAST , UNIQUE_INTENSITY_LEVELS , SHADOW , NOSE_SHADOW , SPECULARITY , EYE_GAZE , RIGHT_EYE_STATUS , LEFT_EYE_STATUS , GLASS_STATUS , HEAVY_FRAME , MOUTH_STATUS , BACKGROUND_UNIFORMITY , AGE , RIGHT_RED_EYE , LEFT_RED_EYE , GENDER , EYE_DISTANCE , ROLL_ANGLE , PITCH_ANGLE , YAW_ANGLE , FACE_SIZE , FACE_RELATIVE_AREA , FACE_RELATIVE_AREA_IN_IMAGE , TINTED_GLASSES , WIDTH_HEIGHT_RATIO }
	Face attributes IDs. More...

Public Member Functions
std::shared_ptr< Keypoints >	GetKeypoints () const
	Provides the facial keypoints detected on the face in the original face capture.
std::shared_ptr< CroppedFace >	Crop (FaceCropConfiguration cfg) const
	Crops the detected face from the original face capture.
Rectangle	DetectionRectangle () const
	Retrieves the bounding box of the detected face.
std::vector< uint8_t >	ToIsoImage (ImageEncoder &encoder) const
	Function creates ISO image based on ISO/IEC 19794-5 (Face image data).
std::shared_ptr< ExtractedFace >	Extract (const FaceExtractorConfig &cfg)
	This function creates a template for the detected face.
std::shared_ptr< ExtractedFace >	Extract ()
	This function creates a template for the detected face with default FaceExtractorConfig.
void	WriteTo (Writer &writer) const final
	Function serializes Face via provided Writer.
bool	IsSameAs (const Face &face) const
	Returns true when argument is same object.
bool	operator== (const Face &f) const
	Compares if Face is equal with other one.
bool	operator!= (const Face &f) const
	Compares if Face is not equal with other one.
float	Sharpness ()
	Sharpness attribute used to measure the sharpness level of an area of the detected face in the original face capture.
float	Brightness ()
	Brightness attribute used to measure the brightness level of an area of the detected face in the original face capture.
float	Contrast ()
	Contrast attribute used to measure the contrast level of an area of the detected face in the original face capture.
float	UniqueIntensityLevels ()
	Unique Intensity Levels attribute used to measure whether an area of the detected face in the original face capture has an appropriate number of unique intensity levels.
float	Shadow ()
	Shadow attribute used to evaluate whether an area of the detected face in the original face capture is affected by shadows.
float	NoseShadow ()
	Nose Shadow attribute for evaluating whether eyes or a nose don't cast sharp shadows.
float	Specularity ()
	Specularity attribute used to evaluate the presence of spotlights in an area of the detected face in the original face capture.
float	EyeGaze ()
	Eye Gaze attribute used to evaluate whether the gaze direction of the detected face in the original face capture is frontal.
float	RightEyeStatus ()
	Attribute for evaluating right eye status.
float	LeftEyeStatus ()
	Attribute for evaluating left eye status.
float	GlassStatus ()
	Attribute for evaluating glasses presence.
float	HeavyFrame ()
	Attribute for evaluating whether glasses with heavy frames are not present.
float	MouthStatus ()
	Attribute for evaluating mouth status.
float	BackgroundUniformity (uint8_t minimalBackgroundRatio)
	Attribute used to measure background uniformity in the close area around the detected face in the original face capture.
float	BackgroundUniformity ()
	Attribute used to measure background uniformity in the close area around the detected face in the original face capture.
float	Age ()
	Attribute for evaluating age of subject using the face.
float	Age (AgeEstimationMode mode)
	Attribute for evaluating age of subject using the face.
float	RightRedEye ()
	Attribute for evaluating whether red-eye effect is not present on right eye.
float	LeftRedEye ()
	Attribute for evaluating whether red-eye effect is not present on left eye.
float	Gender ()
	Attribute for evaluating gender of subject.
float	Gender (GenderEstimationMode mode)
	Attribute for evaluating gender of subject.
float	EyeDistance ()
	Attribute used to measure the distance between the eyes in pixels for the detected face in the original face capture.
float	RollAngle ()
	Attribute representing the head rotation angle around the Z-axis of the detected face in the original face capture, relative to the camera reference frame, as per DIN 9300.
float	PitchAngle ()
	Attribute representing the head rotation angle around the X-axis of the detected face in the original face capture, relative to the camera reference frame, as per DIN 9300.
float	YawAngle ()
	Attribute representing the head rotation angle around the Y-axis of the detected face in the original face capture, relative to the camera reference frame, as per DIN 9300.
float	FaceSize ()
	Attribute representing face size - the maximum of eye distance and eye-mouth distance.
float	FaceRelativeArea ()
	Attribute representing the area of the detected face in the original face capture relative to the size of the original face capture.
float	FaceRelativeAreaInImage ()
	Attribute representing the visible area of the detected face in the original face capture relative to the total face area.
float	TintedGlasses ()
	Attribute for evaluating tinted glasses presence.
float	WidthHeightRatio () const
	Attribute representing width to height aspect ratio of the original face capture.
float	PassiveLiveness ()
	Attribute used to assess whether a detected face in the original face capture qualifies as live, using the PassiveLivenessMode::ACCURATE calculation method.
float	PassiveLiveness (PassiveLivenessMode mode)
	Attribute used to assess whether a detected face in the original face capture qualifies as live.

Public Attributes
const std::shared_ptr< FaceCapture >	capture
	Provides the original FaceCapture.
const unsigned int	confidence
	Provides confidence score of the face related to face detection.

Friends
class	FaceDetector

Detailed Description

It holds all functions and data related to Face manipulation.

Member Enumeration Documentation

◆ AgeEstimationMode

enum class AgeEstimationMode : uint8_t

strong

Specifies the level of compromise between age estimation speed and accuracy.

Enumerator

BALANCED

Provides the fastest processing speed but the lowest age estimation accuracy compared to ACCURATE mode.

Required Solvers

Total size of solvers is 8 MB.

References face-attribute-age-gender-balanced.solver

ACCURATE

Provides better age estimation accuracy compared to BALANCED mode.

However the performance of the age estimation is not as good as when BALANCED mode is used.

Note: The age estimation feature MUST be allowed in license to get accurate age estimation.

Required Solvers

Total size of solvers is 313 MB.

References face-attribute-age-gender-accurate.nnkitsolver

◆ AttributeID

enum class AttributeID : uint8_t

strong

Face attributes IDs.

Enumerator
CONFIDENCE	Provides confidence score of the face related to face detection. See also Face::confidence
SHARPNESS	Sharpness attribute used to measure the sharpness level of an area of the detected face in the original face capture. See also Face::Sharpness
BRIGHTNESS	Brightness attribute used to measure the brightness level of an area of the detected face in the original face capture. See also Face::Brightness
CONTRAST	Contrast attribute used to measure the contrast level of an area of the detected face in the original face capture. See also Face::Contrast
UNIQUE_INTENSITY_LEVELS	Unique Intensity Levels attribute used to measure whether an area of the detected face in the original face capture has an appropriate number of unique intensity levels. See also Face::UniqueIntensityLevels
SHADOW	Shadow attribute used to evaluate whether an area of the detected face in the original face capture is affected by shadows. See also Face::Shadow
NOSE_SHADOW	Nose Shadow attribute for evaluating whether eyes or a nose don't cast sharp shadows. See also Face::NoseShadow
SPECULARITY	Specularity attribute used to evaluate the presence of spotlights in an area of the detected face in the original face capture. See also Face::Specularity
EYE_GAZE	Eye Gaze attribute used to evaluate whether the gaze direction of the detected face in the original face capture is frontal. See also Face::EyeGaze
RIGHT_EYE_STATUS	Attribute for evaluating right eye status. See also Face::RightEyeStatus
LEFT_EYE_STATUS	Attribute for evaluating left eye status. See also Face::LeftEyeStatus
GLASS_STATUS	Attribute for evaluating glasses presence. See also Face::GlassStatus
HEAVY_FRAME	Attribute for evaluating whether glasses with heavy frames are not present. See also Face::HeavyFrame
MOUTH_STATUS	Attribute for evaluating mouth status. See also Face::MouthStatus
BACKGROUND_UNIFORMITY	Attribute used to measure background uniformity in the close area around the detected face in the original face capture. See also Face::BackgroundUniformity
AGE	Attribute for evaluating age of subject using the face. See also Face::Age
RIGHT_RED_EYE	Attribute for evaluating whether red-eye effect is not present on right eye. See also Face::RightRedEye
LEFT_RED_EYE	Attribute for evaluating whether red-eye effect is not present on left eye. See also Face::LeftRedEye
GENDER	Attribute for evaluating gender of subject. See also Face::Gender
EYE_DISTANCE	Attribute used to measure the distance between the eyes in pixels for the detected face in the original face capture. See also Face::EyeDistance
ROLL_ANGLE	Attribute representing the head rotation angle around the Z-axis of the detected face in the original face capture, relative to the camera reference frame, as per DIN 9300. See also Face::RollAngle
PITCH_ANGLE	Attribute representing the head rotation angle around the X-axis of the detected face in the original face capture, relative to the camera reference frame, as per DIN 9300. See also Face::PitchAngle
YAW_ANGLE	Attribute representing the head rotation angle around the Y-axis of the detected face in the original face capture, relative to the camera reference frame, as per DIN 9300. See also Face::YawAngle
FACE_SIZE	Attribute representing face size - the maximum of eye distance and eye-mouth distance. See also Face::FaceSize
FACE_RELATIVE_AREA	Attribute representing the area of the detected face in the original face capture relative to the size of the original face capture. See also Face::FaceRelativeArea
FACE_RELATIVE_AREA_IN_IMAGE	Attribute representing the visible area of the detected face in the original face capture relative to the total face area. See also Face::FaceRelativeAreaInImage
TINTED_GLASSES	Attribute for evaluating tinted glasses presence. See also Face::TintedGlasses
WIDTH_HEIGHT_RATIO	Attribute representing width to height aspect ratio of the original face capture. See also Face::WidthHeightRatio

◆ GenderEstimationMode

enum class GenderEstimationMode : uint8_t

strong

Specifies the level of compromise between age estimation speed and accuracy.

Enumerator

BALANCED

Provides the fastest processing speed but the lowest gender estimation accuracy compared to ACCURATE mode.

Required Solvers

Total size of solvers is 8 MB.

References face-attribute-age-gender-balanced.solver

ACCURATE

Provides better gender estimation accuracy compared to BALANCED mode.

However the performance of the gender estimation is not as good as when BALANCED mode is used.

Note: The gender estimation feature MUST be allowed in license to get accurate gender estimation.

Required Solvers

Total size of solvers is 313 MB.

References face-attribute-age-gender-accurate.nnkitsolver

◆ PassiveLivenessMode

enum class PassiveLivenessMode : uint8_t

strong

Mode of passive liveness calculation.

Enumerator

FAST

Passive liveness mode with best performance available but worse accuracy as ACCURATE mode.

Required Solvers

Total size of solvers is 5 MB.

References face-attribute-passive-liveness-universal-fast.nnkitsolver

ACCURATE

Passive liveness mode with highest accuracy available but significantly slower as FAST mode.

Note: Not supported in this variant of Enrollment SDK.

Member Function Documentation

◆ Age() [1/2]

float Age ( )

inline

Attribute for evaluating age of subject using the face.

It uses BALANCED estimation.

Returns: estimated age

Exceptions

EnrollmentException when attribute can not be calculated

Required Solvers

Total size of solvers is 8 MB.

References face-attribute-age-gender-balanced.solver

References Age(), and BALANCED.

Referenced by Age().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ Age() [2/2]

float Age ( AgeEstimationMode mode )

inline

Attribute for evaluating age of subject using the face.

Parameters

mode	mode of age estimation

Returns: estimated age

Exceptions

EnrollmentException when attribute can not be calculated

Required Solvers: See Face::AgeEstimationMode

◆ BackgroundUniformity() [1/2]

float BackgroundUniformity ( )

inline

Attribute used to measure background uniformity in the close area around the detected face in the original face capture.

The original face capture is stored in the Face::capture member variable.

The attribute can be taken as an ICAO feature.

The default background percentage of the whole image is 5%.

Returns: Background uniformity values are from range <-10000,10000>. Values near -10000 indicates 'very un-uniform background present', values near 10000 indicates 'uniform background present'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

Required Solvers

Total size of solvers is 1 MB.

References face-attribute-background-uniformity.solver

References BackgroundUniformity().

Referenced by BackgroundUniformity().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ BackgroundUniformity() [2/2]

float BackgroundUniformity ( uint8_t minimalBackgroundRatio )

inline

Attribute used to measure background uniformity in the close area around the detected face in the original face capture.

The original face capture is stored in the Face::capture member variable.

The attribute can be taken as an ICAO feature.

Parameters

minimalBackgroundRatio Specifies minimal background percentage of the whole image which needs to be available for background uniformity solver. The values lower as 5% resolves into less accurate result.

Returns: Background uniformity values are from range <-10000,10000>. Values near -10000 indicates 'very un-uniform background present', values near 10000 indicates 'uniform background present'. The decision threshold is around 0.

Exceptions

EnrollmentException	when attribute can not be calculated
EnrollmentInvalidArgumentException	when minimalBackgroundRatio is not in range <0,100>

Required Solvers

Total size of solvers is 1 MB.

References face-attribute-background-uniformity.solver

◆ Brightness()

float Brightness ( )

inline

Brightness attribute used to measure the brightness level of an area of the detected face in the original face capture.

The original face capture is stored in the Face::capture member variable.

The attribute can be taken as an ICAO feature.

Returns: Brightness values are from range <-10000,10000>. Values near -10000 indicates 'too dark', values near 10000 indicates 'too light', values around 0 indicates OK. The decision thresholds are around -5000 and 5000.

Exceptions

EnrollmentException when attribute can not be calculated

◆ Contrast()

float Contrast ( )

inline

Contrast attribute used to measure the contrast level of an area of the detected face in the original face capture.

The original face capture is stored in the Face::capture member variable.

The attribute can be taken as an ICAO feature.

Returns: Contrast values are from range <-10000,10000>. Values near -10000 indicates 'very low contrast', values near 10000 indicates 'very high contrast', values around 0 indicates OK. The decision thresholds are around -5000 and 5000.

Exceptions

EnrollmentException when attribute can not be calculated

◆ Crop()

std::shared_ptr< CroppedFace > Crop ( FaceCropConfiguration cfg ) const

inlinenodiscard

Crops the detected face from the original face capture.

This method extracts a cropped version of the detected face based on the given cropping configuration. The original face capture is stored in the Face::capture member variable.

Parameters

cfg	The configuration specifying how the face should be cropped.

Returns: CroppedFace object containing the cropped face.

References capture.

◆ DetectionRectangle()

Rectangle DetectionRectangle ( ) const

inlinenodiscard

Retrieves the bounding box of the detected face.

This method returns the rectangular region that encloses the detected face within the original face capture.

Returns: A Rectangle representing the bounding box of the detected face.

◆ Extract() [1/2]

std::shared_ptr< ExtractedFace > Extract ( )

inline

This function creates a template for the detected face with default FaceExtractorConfig.

Returns: template calculated template

Exceptions

EnrollmentException when it is not possible to extract templates

References FaceExtractorConfig::Default(), and Extract().

Referenced by Extract().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ Extract() [2/2]

std::shared_ptr< ExtractedFace > Extract ( const FaceExtractorConfig & cfg )

inline

This function creates a template for the detected face.

Parameters

cfg	FaceExtractorConfig

Returns: template calculated template

Exceptions

EnrollmentException when it is not possible to extract templates

Required Solvers: See FaceExtractorConfig::Type

References capture, and FaceExtractorConfig::GetExtractor().

Here is the call graph for this function:

◆ EyeDistance()

float EyeDistance ( )

inline

Attribute used to measure the distance between the eyes in pixels for the detected face in the original face capture.

The original face capture is stored in the Face::capture member variable.

The attribute can be taken as an ICAO feature.

Returns: Distance in pixels.

Exceptions

EnrollmentException when attribute can not be calculated

Required Solvers: See FaceDetector::Config::Type

◆ EyeGaze()

float EyeGaze ( )

inline

Eye Gaze attribute used to evaluate whether the gaze direction of the detected face in the original face capture is frontal.

The original face capture is stored in the Face::capture member variable.

The attribute can be taken as an ICAO feature.

Returns: Eye gaze values are from range <-10000,10000>. Values near -10000 indicates 'sideway ahead eyes gaze', values near 10000 indicates 'straight ahead eyes gaze'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

Required Solvers

Total size of solvers is 944 kB.

References face-attribute-eye-gaze.solver

◆ FaceRelativeArea()

float FaceRelativeArea ( )

inline

Attribute representing the area of the detected face in the original face capture relative to the size of the original face capture.

The original face capture is stored in the Face::capture member variable.

Returns: area of face relative to image size in percentage. It can be more that 100% due to borders added by face detector.

Exceptions

EnrollmentException when attribute can not be calculated

◆ FaceRelativeAreaInImage()

float FaceRelativeAreaInImage ( )

inline

Attribute representing the visible area of the detected face in the original face capture relative to the total face area.

This value indicates the percentage of the face area that is outside the image. The original face capture is stored in the Face::capture member variable.

Returns: area of face visible in image relative to total area of face. This value implies the percentage of face area outside the image.

Exceptions

EnrollmentException when attribute can not be calculated

◆ FaceSize()

float FaceSize ( )

inline

Attribute representing face size - the maximum of eye distance and eye-mouth distance.

Returns: face size - the maximum of eye distance and eye-mouth distance in pixels.

Exceptions

EnrollmentException when attribute can not be calculated

◆ Gender() [1/2]

float Gender ( )

inline

Attribute for evaluating gender of subject.

It uses BALANCED estimation.

Returns: Gender values are from range <-10000, 10000>. Values near -10000 indicates 'male', values near 10000 indicates 'female'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

Required Solvers

Total size of solvers is 8 MB.

References face-attribute-age-gender-balanced.solver

References BALANCED, and Gender().

Referenced by Gender().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ Gender() [2/2]

float Gender ( GenderEstimationMode mode )

inline

Attribute for evaluating gender of subject.

Parameters

mode	mode of gender estimation

Returns: Gender values are from range <-10000, 10000>. Values near -10000 indicates 'male', values near 10000 indicates 'female'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

Required Solvers

Total size of solvers is 8 MB.

References face-attribute-age-gender-balanced.solver

◆ GetKeypoints()

std::shared_ptr< Keypoints > GetKeypoints ( ) const

inlinenodiscard

Provides the facial keypoints detected on the face in the original face capture.

The original face capture is stored in the Face::capture member variable.

Returns: Keypoints instance.

◆ GlassStatus()

float GlassStatus ( )

inline

Attribute for evaluating glasses presence.

Returns: Glasses values are from range <-10000,10000>. Values near -10000 indicates 'no glasses present', values near 10000 indicates 'glasses present'. The decision threshold is around 0

Exceptions

EnrollmentException when attribute can not be calculated

Required Solvers

Total size of solvers is 608 kB.

References face-attribute-glasses.solver

◆ HeavyFrame()

float HeavyFrame ( )

inline

Attribute for evaluating whether glasses with heavy frames are not present.

The attribute can be taken as an ICAO feature.

Returns: Heavy frame glasses values are from range <-10000,10000>. Values near -10000 indicates 'no heavy frame glasses present', values near 10000 indicates 'heavy frame glasses present'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

Required Solvers

Total size of solvers is 498 kB.

References face-attribute-heavy-frame.solver

◆ IsSameAs()

bool IsSameAs ( const Face & face ) const

inline

Returns true when argument is same object.

Parameters

face	to check if it is reference to same object

Returns: true when given Face is the same object; false when given Face is a different object

◆ LeftEyeStatus()

float LeftEyeStatus ( )

inline

Attribute for evaluating left eye status.

The attribute can be taken as an ICAO feature.

Returns: Left eye values are from range <-10000,10000>. Values near -10000 indicates 'closed, narrowed or bulged eye', values near 10000 indicates 'normally opened eye'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

Required Solvers

Total size of solvers is 16 kB.

References face-attribute-eye-status.solver

◆ LeftRedEye()

float LeftRedEye ( )

inline

Attribute for evaluating whether red-eye effect is not present on left eye.

The attribute can be taken as an ICAO feature.

Returns: Left red-eye effect values are from range <-10000,10000>. Values near -10000 indicates 'no red-eye effect present', values near 10000 indicates 'red-eye effect present'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

◆ MouthStatus()

float MouthStatus ( )

inline

Attribute for evaluating mouth status.

The attribute can be taken as an ICAO feature.

Returns: Mouth status values are from range <-10000,10000>. Values near -10000 indicates 'open mouth, smile showing teeth or round lips present', values near 10000 indicates 'mouth with no expression'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

Required Solvers

Total size of solvers is 155 kB.

References face-attribute-mouth-status.solver

◆ NoseShadow()

float NoseShadow ( )

inline

Nose Shadow attribute for evaluating whether eyes or a nose don't cast sharp shadows.

The attribute can be taken as an ICAO feature.

Returns: Nose shadow values are from range <-10000,10000> Values near -10000 indicates 'very strong local (eyes/nose) shadows present', values near 10000 indicates 'no local shadows present'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

◆ operator!=()

bool operator!= ( const Face & f ) const

inline

Compares if Face is not equal with other one.

Parameters

f	compared Face

Returns: true when data of Face are different; false when data of Face are same

◆ operator==()

bool operator== ( const Face & f ) const

inline

Compares if Face is equal with other one.

Parameters

f	compared Face

Returns: true when data of Face are same; false when data of Face are different

References capture.

◆ PassiveLiveness() [1/2]

float PassiveLiveness ( )

inline

Attribute used to assess whether a detected face in the original face capture qualifies as live, using the PassiveLivenessMode::ACCURATE calculation method.

The original face capture is stored in the Face::capture member variable.

Returns: Passive Liveness score from range <0,100>.

Note: Not supported in this variant of Enrollment SDK.

References ACCURATE, and PassiveLiveness().

Referenced by PassiveLiveness().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ PassiveLiveness() [2/2]

float PassiveLiveness ( PassiveLivenessMode mode )

inline

Attribute used to assess whether a detected face in the original face capture qualifies as live.

The original face capture is stored in the Face::capture member variable.

Exceptions

EnrollmentException when attribute can not be calculated

Parameters

mode	mode of passive liveness calculation. For more information see Face Passive Liveness.

Note: The passive liveness feature MUST be allowed in license to get face passive liveness.

Returns: Passive Liveness score from range <0,100>.; When mode is PassiveLivenessMode::FAST then Fast Passive Liveness Thresholds can be used to check live and non-live faces.

Required Solvers: See Face::PassiveLivenessMode

References FAST.

◆ PitchAngle()

float PitchAngle ( )

inline

Attribute representing the head rotation angle around the X-axis of the detected face in the original face capture, relative to the camera reference frame, as per DIN 9300.

The original face capture is stored in the Face::capture member variable.

Returns: angle rotation of head towards camera reference frame around X-axis as per DIN9300.

Exceptions

EnrollmentException when attribute can not be calculated

◆ RightEyeStatus()

float RightEyeStatus ( )

inline

Attribute for evaluating right eye status.

The attribute can be taken as an ICAO feature.

Returns: Right eye values are from range <-10000,10000>. Values near -10000 indicates 'closed, narrowed or bulged eye', values near 10000 indicates 'normally opened eye'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

Required Solvers

Total size of solvers is 16 kB.

References face-attribute-eye-status.solver

◆ RightRedEye()

float RightRedEye ( )

inline

Attribute for evaluating whether red-eye effect is not present on right eye.

The attribute can be taken as an ICAO feature.

Returns: Right red-eye effect values are from range <-10000,10000>. Values near -10000 indicates 'no red-eye effect present', values near 10000 indicates 'red-eye effect present'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

◆ RollAngle()

float RollAngle ( )

inline

Attribute representing the head rotation angle around the Z-axis of the detected face in the original face capture, relative to the camera reference frame, as per DIN 9300.

The original face capture is stored in the Face::capture member variable.

Returns: angle rotation of head towards camera reference frame around Z-axis as per DIN9300.

Exceptions

EnrollmentException when attribute can not be calculated

◆ Shadow()

float Shadow ( )

inline

Shadow attribute used to evaluate whether an area of the detected face in the original face capture is affected by shadows.

The original face capture is stored in the Face::capture member variable.

The attribute can be taken as an ICAO feature.

Returns: Shadow values are from range <-10000,10000>. Values near -10000 indicates 'very strong global shadows present', values near 10000 indicates 'no global shadows present'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

◆ Sharpness()

float Sharpness ( )

inline

Sharpness attribute used to measure the sharpness level of an area of the detected face in the original face capture.

The original face capture is stored in the Face::capture member variable.

The attribute can be taken as an ICAO feature.

Returns: Sharpness values are from range <-10000,10000>. Values near -10000 indicates 'very blurred', values near 10000 indicates 'very sharp'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

◆ Specularity()

float Specularity ( )

inline

Specularity attribute used to evaluate the presence of spotlights in an area of the detected face in the original face capture.

The original face capture is stored in the Face::capture member variable.

The attribute can be taken as an ICAO feature.

Returns: Specularity values are from range <-10000,10000> Values near -10000 indicates 'very strong specularity present', values near 10000 indicates 'no specularity present'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

◆ TintedGlasses()

float TintedGlasses ( )

inline

Attribute for evaluating tinted glasses presence.

Returns: Tinted glasses values are from range <-10000,10000>. Values near -10000 indicates 'no tinted glasses present', values near 10000 indicates 'tinted glasses present'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

Required Solvers

Total size of solvers is 608 kB.

References face-attribute-glasses.solver

◆ ToIsoImage()

std::vector< uint8_t > ToIsoImage ( ImageEncoder & encoder ) const

inlinenodiscard

Function creates ISO image based on ISO/IEC 19794-5 (Face image data).

ISO standard defines a standard data format of digital face images to allow interoperability among vendors.

Parameters

encoder specific encoder for encoding Print image. Supported JPEG 2000 (lossy & lossless), PNG

Returns: ISO image from ISO/IEC 19794-5 (Face image data)

Exceptions

ImageException	when it is not possible to encode image
EnrollmentInvalidArgumentException	when encoder is not supported
EnrollmentRuntimeException	for invalid conversion size

◆ UniqueIntensityLevels()

float UniqueIntensityLevels ( )

inline

Unique Intensity Levels attribute used to measure whether an area of the detected face in the original face capture has an appropriate number of unique intensity levels.

The original face capture is stored in the Face::capture member variable.

The attribute can be taken as an ICAO feature.

Returns: Unique intensity levels values are from range <-10000,10000>. Values near -10000 indicates 'very few unique intensity levels', values near 10000 indicates 'enough unique intensity levels'. The decision threshold is around 0.

Exceptions

EnrollmentException when attribute can not be calculated

◆ WidthHeightRatio()

float WidthHeightRatio ( ) const

inline

Attribute representing width to height aspect ratio of the original face capture.

This is calculated as an image width divided image height, as defined in B.3.2.2 Width to height ratio of the image [6].

The original face capture is stored in the Face::capture member variable.

Returns: width to height aspect ratio of the original Face::capture

◆ WriteTo()

void WriteTo ( Writer & writer ) const

inlinefinalvirtual

Function serializes Face via provided Writer.

Parameters

writer used to serialize Face.

Implements Serializable.

References capture, and confidence.

◆ YawAngle()

float YawAngle ( )

inline

Attribute representing the head rotation angle around the Y-axis of the detected face in the original face capture, relative to the camera reference frame, as per DIN 9300.

The original face capture is stored in the Face::capture member variable.

Returns: angle rotation of head towards camera reference frame around Y-axis as per DIN9300.

Exceptions

EnrollmentException when attribute can not be calculated

Member Data Documentation

◆ capture

const std::shared_ptr<FaceCapture> capture

Provides the original FaceCapture.

It contains the image in which this face was detected.

Referenced by Crop(), Extract(), operator==(), and WriteTo().

◆ confidence

const unsigned int confidence

Provides confidence score of the face related to face detection.

It can be taken as an ICAO feature. Confidence values are from range <0,10000>. The higher the value of the attribute the better quality of the face. The decision thresholds are around 600, but it depends on the face image quality / camera angle etc.

Referenced by WriteTo().

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

Face/Face.h

Classes

Public Types

Public Member Functions

Public Attributes

Friends

Detailed Description

Member Enumeration Documentation

◆ AgeEstimationMode

◆ AttributeID

◆ GenderEstimationMode

◆ PassiveLivenessMode

Member Function Documentation

◆ Age() [1/2]

◆ Age() [2/2]

◆ BackgroundUniformity() [1/2]

◆ BackgroundUniformity() [2/2]

◆ Brightness()

◆ Contrast()

◆ Crop()

◆ DetectionRectangle()

◆ Extract() [1/2]

◆ Extract() [2/2]

◆ EyeDistance()

◆ EyeGaze()

◆ FaceRelativeArea()

◆ FaceRelativeAreaInImage()

◆ FaceSize()

◆ Gender() [1/2]

◆ Gender() [2/2]

◆ GetKeypoints()

◆ GlassStatus()

◆ HeavyFrame()

◆ IsSameAs()

◆ LeftEyeStatus()

◆ LeftRedEye()

◆ MouthStatus()

◆ NoseShadow()

◆ operator!=()

◆ operator==()

◆ PassiveLiveness() [1/2]

◆ PassiveLiveness() [2/2]

◆ PitchAngle()

◆ RightEyeStatus()

◆ RightRedEye()

◆ RollAngle()

◆ Shadow()

◆ Sharpness()

◆ Specularity()

◆ TintedGlasses()

◆ ToIsoImage()

◆ UniqueIntensityLevels()

◆ WidthHeightRatio()

◆ WriteTo()

◆ YawAngle()

Member Data Documentation

◆ capture

◆ confidence