Vision Face

Face

Face class representing the Vision API’s face detection response.

class google.cloud.vision.face.Angles(roll, pan, tilt)

Bases: object

Angles representing the positions of a face.

classmethod from_api_repr(response)

Factory: construct the angles from an Vision API response.

Return type:Angles Returns:An Angles instance with data parsed from response.

pan

Pan angle of face.

Return type:float Returns:Pan angle of face in degrees.

roll

Roll angle of face.

Return type:float Returns:Roll angle of face in degrees.

tilt

Tilt angle of face.

Return type:float Returns:Tilt angle of face in degrees.

class google.cloud.vision.face.Bounds(vertices)

Bases: google.cloud.vision.geometry.BoundsBase

The bounding polygon of the entire face.

class google.cloud.vision.face.Emotions(joy_likelihood, sorrow_likelihood, surprise_likelihood, anger_likelihood)

Bases: object

Emotions displayed by the face detected in an image.

anger_likelihood

Likelihood of anger in detected face.

Return type:str Returns:String derived from Likelihood.

classmethod from_api_repr(response)

Factory: construct Emotions from Vision API response.

Parameters:response (dict) – Response dictionary representing a face. Return type:Emotions Returns:Populated instance of Emotions.

joy_likelihood

Likelihood of joy in detected face.

Return type:str Returns:String derived from Likelihood.

sorrow_likelihood

Likelihood of sorrow in detected face.

Return type:str Returns:String derived from Likelihood.

surprise_likelihood

Likelihood of surprise in detected face.

Return type:str Returns:String derived from Likelihood.

class google.cloud.vision.face.FDBounds(vertices)

Bases: google.cloud.vision.geometry.BoundsBase

The bounding polygon of just the skin portion of the face.

class google.cloud.vision.face.Face(angles, bounds, detection_confidence, emotions, fd_bounds, headwear_likelihood, image_properties, landmarks, landmarking_confidence)

Bases: object

Representation of a face found by the Vision API

angles

Accessor to the pan, tilt and roll angles of a Face.

Return type:Angles Returns:Pan, tilt and roll angles of the detected face.

bounds

Accessor to the bounding poly information of the detected face.

Return type:Bounds Returns:An instance of Bounds which has a list of vertices.

detection_confidence

Face detection confidence score determined by the Vision API.

Return type:float Returns:Float representation of confidence ranging from 0 to 1.

emotions

Accessor to the possible emotions expressed in the detected face.

Return type:Emotions Returns:An instance of Emotions with joy, sorrow, anger, surprise likelihood.

fd_bounds

Accessor to the skin area bounding poly of the detected face.

Return type:FDBounds Returns:An instance of FDBounds which has a list of vertices.

classmethod from_api_repr(response)

Factory: construct an instance of a Face from an API response

Parameters:response (dict) – Face annotation dict returned from the Vision API. Return type:Face Returns:A instance of Face with data parsed from response.

headwear_likelihood

Headwear likelihood.

Return type:Likelihood Returns:String representing the likelihood based on Likelihood

image_properties

Image properties from imaged used in face detection.

Return type:FaceImageProperties Returns:FaceImageProperties object with image properties.

landmarking_confidence

Landmarking confidence score determinged by the Vision API.

Return type:float Returns:Float representing the confidence of the Vision API in determining the landmarks on a face.

landmarks

Accessor to the facial landmarks detected in a face.

Return type:Landmarks Returns:Landmarks object with facial landmarks as properies.

class google.cloud.vision.face.FaceImageProperties(blurred_likelihood, underexposed_likelihood)

Bases: object

A representation of the image properties from face detection.

blurred_likelihood

Likelihood of the image being blurred.

Return type:str Returns:String representation derived from Position.

classmethod from_api_repr(response)

Factory: construct image properties from image.

Return type:FaceImageProperties Returns:Instance populated with image property data.

underexposed_likelihood

Likelihood that the image used for detection was underexposed.

Return type:str Returns:String representation derived from Position.

class google.cloud.vision.face.FaceLandmarkTypes

Bases: object

A representation of the face detection landmark types.

See: https://cloud.google.com/vision/reference/rest/v1/images/annotate#Type_1

CHIN_GNATHION = 'CHIN_GNATHION'

CHIN_LEFT_GONION = 'CHIN_LEFT_GONION'

CHIN_RIGHT_GONION = 'CHIN_RIGHT_GONION'

FOREHEAD_GLABELLA = 'FOREHEAD_GLABELLA'

LEFT_EAR_TRAGION = 'LEFT_EAR_TRAGION'

LEFT_EYE = 'LEFT_EYE'

LEFT_EYEBROW_UPPER_MIDPOINT = 'LEFT_EYEBROW_UPPER_MIDPOINT'

LEFT_EYE_BOTTOM_BOUNDARY = 'LEFT_EYE_BOTTOM_BOUNDARY'

LEFT_EYE_LEFT_CORNER = 'LEFT_EYE_LEFT_CORNER'

LEFT_EYE_PUPIL = 'LEFT_EYE_PUPIL'

LEFT_EYE_RIGHT_CORNER = 'LEFT_EYE_RIGHT_CORNER'

LEFT_EYE_TOP_BOUNDARY = 'LEFT_EYE_TOP_BOUNDARY'

LEFT_OF_LEFT_EYEBROW = 'LEFT_OF_LEFT_EYEBROW'

LEFT_OF_RIGHT_EYEBROW = 'LEFT_OF_RIGHT_EYEBROW'

LOWER_LIP = 'LOWER_LIP'

MIDPOINT_BETWEEN_EYES = 'MIDPOINT_BETWEEN_EYES'

MOUTH_CENTER = 'MOUTH_CENTER'

MOUTH_LEFT = 'MOUTH_LEFT'

MOUTH_RIGHT = 'MOUTH_RIGHT'

NOSE_BOTTOM_CENTER = 'NOSE_BOTTOM_CENTER'

NOSE_BOTTOM_LEFT = 'NOSE_BOTTOM_LEFT'

NOSE_BOTTOM_RIGHT = 'NOSE_BOTTOM_RIGHT'

NOSE_TIP = 'NOSE_TIP'

RIGHT_EAR_TRAGION = 'RIGHT_EAR_TRAGION'

RIGHT_EYE = 'RIGHT_EYE'

RIGHT_EYEBROW_UPPER_MIDPOINT = 'RIGHT_EYEBROW_UPPER_MIDPOINT'

RIGHT_EYE_BOTTOM_BOUNDARY = 'RIGHT_EYE_BOTTOM_BOUNDARY'

RIGHT_EYE_LEFT_CORNER = 'RIGHT_EYE_LEFT_CORNER'

RIGHT_EYE_PUPIL = 'RIGHT_EYE_PUPIL'

RIGHT_EYE_RIGHT_CORNER = 'RIGHT_EYE_RIGHT_CORNER'

RIGHT_EYE_TOP_BOUNDARY = 'RIGHT_EYE_TOP_BOUNDARY'

RIGHT_OF_LEFT_EYEBROW = 'RIGHT_OF_LEFT_EYEBROW'

RIGHT_OF_RIGHT_EYEBROW = 'RIGHT_OF_RIGHT_EYEBROW'

UNKNOWN_LANDMARK = 'UNKNOWN_LANDMARK'

UPPER_LIP = 'UPPER_LIP'

class google.cloud.vision.face.Landmark(position, landmark_type)

Bases: object

A face-specific landmark (for example, a face feature, left eye).

classmethod from_api_repr(response_landmark)

Factory: construct an instance of a Landmark from a response.

Parameters:response_landmark (dict) – Landmark representation from Vision API. Return type:Landmark Returns:Populated instance of Landmark.

landmark_type

Landmark type of facial feature.

Return type:str Returns:String representation of facial landmark type.

position

Landmark position on face.

Return type:Position Returns:Instance of Position with landmark coordinates.

class google.cloud.vision.face.Landmarks(landmarks)

Bases: object

Landmarks detected on a face represented as properties.