def estimate(self, warp: Union[FaceWarp, FaceWarpedImage],
estimateAge: bool, estimateGender: bool,
estimateEthnicity: bool) -> BasicAttributes:
"""
Estimate a basic attributes (age, gender, ethnicity) from warped images.
Args:
warp: warped image
estimateAge: estimate age or not
estimateGender: estimate gender or not
estimateEthnicity: estimate ethnicity or not
Returns:
estimated age, gender, ethnicity
Raises:
LunaSDKException: if estimation failed
"""
dtAttributes = 0
if estimateAge:
dtAttributes |= AttributeRequest.estimateAge
if estimateGender:
dtAttributes |= AttributeRequest.estimateGender
if estimateEthnicity:
dtAttributes |= AttributeRequest.estimateEthnicity
error, baseAttributes = self._coreEstimator.estimate(
warp.warpedImage.coreImage, AttributeRequest(dtAttributes))
assertError(error)
return BasicAttributes(baseAttributes)
def estimateBatch(
self, images: List[Union[VLImage, FaceWarp, FaceWarpedImage]]
) -> List[OrientationType]:
"""
Batch estimate orientation mode from warped images.
Args:
images: vl image or face warp list
Returns:
estimated orientation mode list
Raises:
LunaSDKException: if estimation is failed
"""
coreImages = [
img.warpedImage.coreImage
if isinstance(img, FaceWarp) else img.coreImage for img in images
]
validateInputByBatchEstimator(self._coreEstimator, coreImages)
error, coreOrientationTypeList = self._coreEstimator.estimate(
coreImages)
assertError(error)
return [
OrientationType.fromCoreOrientationType(coreOrientationType)
for coreOrientationType in coreOrientationTypeList
]
def postProcessingBatchWithAggregation(
error: FSDKErrorResult,
coreEstimations: List[CoreTaskResult],
aggregatedAttribute: CoreTaskResult,
resultClass: Type[CommonTaskResult],
aggregate: bool,
) -> Tuple[List[CommonTaskResult], Optional[CommonTaskResult]]:
"""
Post processing batch estimation
Args:
error: estimation error
coreEstimations: core batch estimation
aggregatedAttribute: aggregated core estimation
aggregate: need or not aggregate result
resultClass: result class
Returns:
list of lunavl structure based on core estimation + aggregated estimation
"""
assertError(error)
estimations = [
resultClass(coreEstimation) for coreEstimation in coreEstimations
]
if aggregate:
return estimations, resultClass(aggregatedAttribute)
else:
return estimations, None
def estimateBatch(self, warpWithLandmarksList: List[WarpWithLandmarks]) -> List[EyesEstimation]:
"""
Batch estimate mouth state on warps.
Args:
warpWithLandmarksList: list of core warp with transformed landmarks
Returns:
list of estimated states
Raises:
LunaSDKException: if estimation failed
ValueError: if warps count not equals landmarks count
"""
cropper = EyeCropper()
eyeRectList = []
for warpWithLandmarks in warpWithLandmarksList:
if isinstance(warpWithLandmarks.landmarks, Landmarks5):
eyeRectList.append(
cropper.cropByLandmarks5(
warpWithLandmarks.warp.warpedImage.coreImage, warpWithLandmarks.landmarks.coreEstimation
)
)
else:
eyeRectList.append(
cropper.cropByLandmarks68(
warpWithLandmarks.warp.warpedImage.coreImage, warpWithLandmarks.landmarks.coreEstimation
)
)
coreImages = [row.warp.warpedImage.coreImage for row in warpWithLandmarksList]
validateInputByBatchEstimator(self._coreEstimator, coreImages, eyeRectList)
error, eyesEstimations = self._coreEstimator.estimate(coreImages, eyeRectList)
assertError(error)
return [EyesEstimation(eyesEstimation) for eyesEstimation in eyesEstimations]
def makeWarpTransformationWithLandmarks(
self, faceDetection: FaceDetection,
typeLandmarks: str) -> Union[Landmarks68, Landmarks5]:
"""
Make warp transformation with landmarks
Args:
faceDetection: face detection with landmarks5
typeLandmarks: landmarks for warping ("L68" or "L5")
Returns:
warping landmarks
Raises:
ValueError: if landmarks5 is not estimated
LunaSDKException: if transform failed
"""
transformation = self._createWarpTransformation(faceDetection)
if typeLandmarks == "L68":
if faceDetection.landmarks68 is None:
raise ValueError("landmarks68 does not estimated")
error, warp = self._coreWarper.warp(
faceDetection.landmarks68.coreEstimation, transformation)
elif typeLandmarks == "L5":
if faceDetection.landmarks5 is None:
raise ValueError("landmarks5 does not estimated")
error, warp = self._coreWarper.warp(
faceDetection.landmarks5.coreEstimation, transformation)
else:
raise ValueError(
"Invalid value of typeLandmarks, must be 'L68' or 'L5'")
assertError(error)
if typeLandmarks == "L68":
return Landmarks68(warp)
return Landmarks5(warp)
def postProcessingBatchWithAggregation(
error: FSDKErrorResult,
aggregetionGs: float,
gScores: List[float],
descriptorBatch: GenericDescriptorBatch,
aggregatedDescriptor: GenericDesciptor,
) -> Tuple[GenericDescriptorBatch, GenericDesciptor]:
"""
Post processing batch extraction result with aggregation, error check.
Args:
error: extractor error, usually error.isError is False
gScores: garbage scores of extracted descriptors
descriptorBatch: extracted descriptor batch
aggregetionGs: garbage score of aggregated descriptor
aggregatedDescriptor: aggregated descriptor
Raises:
LunaSDKException: if extraction is failed
Returns:
descriptor batch + aggregated descriptor
"""
assertError(error)
aggregatedDescriptor.garbageScore = aggregetionGs
descriptorBatch.scores = gScores
return descriptorBatch, aggregatedDescriptor
def estimate(
self,
detection: Optional[FaceDetection] = None,
imageWithFaceDetection: Optional[ImageWithFaceDetection] = None
) -> float:
"""
Estimate ags for single image/detection.
Args:
detection: face detection
imageWithFaceDetection: image with face detection
Returns:
estimated ags, float in range[0,1]
Raises:
LunaSDKException: if estimation failed
ValueError: if image and detection are None
"""
if detection is None:
if imageWithFaceDetection is None:
raise ValueError(
"image and boundingBox or detection must be not None")
error, ags = self._coreEstimator.estimate(
imageWithFaceDetection.image.coreImage,
imageWithFaceDetection.boundingBox.coreEstimation)
else:
error, ags = self._coreEstimator.estimate(
detection.image.coreImage,
detection.boundingBox.coreEstimation)
assertError(error)
return ags
def estimateBatch(
self, detections: Union[List[FaceDetection],
List[ImageWithFaceDetection]]
) -> List[float]:
"""
Estimate ags for list of detections.
Args:
detections: face detection list or list of image with its face detection
Returns:
list of estimated ags, float in range[0,1]
Raises:
LunaSDKException: if estimation failed
ValueError: if empty image list and empty detection list or images count not match bounding boxes count
"""
coreImages = [detection.image.coreImage for detection in detections]
boundingBoxEstimations = [
detection.boundingBox.coreEstimation for detection in detections
]
validateInputByBatchEstimator(self._coreEstimator, coreImages,
boundingBoxEstimations)
error, agsList = self._coreEstimator.estimate(coreImages,
boundingBoxEstimations)
assertError(error)
return agsList
def estimate( # type: ignore
self,
faceDetection: FaceDetection,
qualityThreshold: Optional[float] = None) -> LivenessV1:
"""
Estimate a liveness
.. warning::
Current estimator version estimates correct liveness state for images from mobile and web camera only.
A correctness of a liveness prediction are not guarantee for other images source.
Args:
faceDetection: face detection
qualityThreshold: quality threshold. if estimation quality is low of this threshold
Returns:
estimated liveness
Raises:
LunaSDKException: if estimation failed
"""
if faceDetection.landmarks5 is None:
raise ValueError("Landmarks5 is required for liveness estimation")
error, estimation = self._coreEstimator.estimate(
faceDetection.image.coreImage,
faceDetection.coreEstimation.detection,
faceDetection.landmarks5.coreEstimation,
-1.0 if qualityThreshold is None else qualityThreshold,
)
assertError(error)
prediction = LivenessPrediction.fromCoreEmotion(estimation.State)
return LivenessV1(estimation, prediction)
def estimate(
warp: Union[HumanWarp, HumanWarpedImage, FaceWarp, FaceWarpedImage],
descriptorFactory: BaseDescriptorFactory,
coreEstimator: IDescriptorExtractorPtr,
descriptor: Optional[BaseDescriptor] = None,
) -> BaseDescriptor:
"""
Estimate a face descriptor or a human descriptor from the warped image.
Args:
warp: warped image
descriptor: descriptor for saving extract result
descriptorFactory: descriptor factory
coreEstimator: descriptor extractor
Returns:
estimated descriptor
Raises:
LunaSDKException: if estimation failed
"""
if descriptor is None:
descriptor = descriptorFactory.generateDescriptor()
coreDescriptor = descriptor.coreEstimation
else:
coreDescriptor = descriptor.coreEstimation
error, optionalGS = coreEstimator.extractFromWarpedImage(
warp.warpedImage.coreImage, coreDescriptor)
assertError(error)
descriptor.garbageScore = optionalGS
return descriptor
def append(self, descriptor: FaceDescriptor) -> None:
"""
Appends descriptor to internal storage.
Args:
descriptor: descriptor with correct length, version and data
Raises:
LunaSDKException: if an error occurs while adding the descriptor
"""
error = self._coreIndex.appendDescriptor(descriptor.coreEstimation)
assertError(error)
def appendBatch(self, descriptorsBatch: FaceDescriptorBatch) -> None:
"""
Appends batch of descriptors to internal storage.
Args:
descriptorsBatch: batch of descriptors with correct length, version and data
Raises:
LunaSDKException: if an error occurs while adding the batch of descriptors
"""
error = self._coreIndex.appendBatch(descriptorsBatch.coreEstimation)
assertError(error)
def buildIndex(self) -> DynamicIndex:
"""
Build index with all appended descriptors.
Raises:
LunaSDKException: if an error occurs while building the index
Returns:
DynamicIndex
"""
error, index = self._coreIndex.buildIndex()
assertError(error)
return DynamicIndex(index, self._faceEngine)
def __delitem__(self, index: int) -> None:
"""
Descriptor will be removed from the graph (not from the internal storage), so it is not available for search.
Args:
index: identification of descriptors position in internal storage
Raises:
IndexError: if index out of range
LunaSDKException: if an error occurs while remove descriptor failed
"""
error = self._coreIndex.removeDescriptor(index)
assertError(error)
def match(
self,
reference: Union[FaceDescriptor, bytes],
candidates: Union[FaceDescriptor, bytes, List[Union[FaceDescriptor,
bytes]],
FaceDescriptorBatch],
) -> Union[MatchingResult, List[MatchingResult]]:
"""
Match face descriptor vs face descriptors.
Returns:
List of matching results if match by several descriptors otherwise one MatchingResult.
"""
if isinstance(reference, bytes):
referenceForMatcher = self.descriptorFactory.generateDescriptor(
reference)
else:
referenceForMatcher = reference
if isinstance(candidates, bytes):
candidatesForMatcher = self.descriptorFactory.generateDescriptor(
candidates)
elif isinstance(candidates, list):
candidatesForMatcher = []
for idx in range(len(candidates)):
if isinstance(candidates[idx], bytes):
candidatesForMatcher.append(
self.descriptorFactory.generateDescriptor(
candidates[idx]))
else:
candidatesForMatcher.append(candidates[idx])
else:
candidatesForMatcher = candidates
if isinstance(candidatesForMatcher, FaceDescriptor):
error, matchResults = self._coreMatcher.match(
referenceForMatcher.coreEstimation,
candidatesForMatcher.coreEstimation)
elif isinstance(candidatesForMatcher, FaceDescriptorBatch):
error, matchResults = self._coreMatcher.match(
referenceForMatcher.coreEstimation,
candidatesForMatcher.coreEstimation)
else:
batch = self.descriptorFactory.generateDescriptorsBatch(
len(candidatesForMatcher))
for candidate in candidatesForMatcher:
batch.append(candidate)
error, matchResults = self._coreMatcher.match(
referenceForMatcher.coreEstimation, batch.coreEstimation)
assertError(error)
return matchResults
def postProcessing(error: FSDKErrorResult, orientationType):
"""
Postprocessing single core image orientation estimation
Args:
error: estimation error
orientationType: core estimation
Returns:
image orientation
"""
assertError(error)
return OrientationType.fromCoreOrientationType(orientationType)
def postProcessing(error: FSDKErrorResult, coreEstimation: CoreTaskResult,
resultClass: Type[CommonTaskResult]) -> CommonTaskResult:
"""
Postprocessing single core estimation
Args:
error: estimation error
coreEstimation: core estimation
Returns:
list of lunavl structure based on core estimation
"""
assertError(error)
return resultClass(coreEstimation)
def _getDynamicIndex(self, path: str) -> DynamicIndex:
"""
Get dynamic index from file
Args:
path: path to saved index
Raises:
LunaSDKException: if an error occurs while loading the index
Returns:
dynamic index
"""
error, loadedIndex = self._faceEngine.loadDynamicIndex(path)
assertError(error)
return DynamicIndex(loadedIndex, self._faceEngine)
def postProcessing(error: FSDKErrorResult,
estimation: LivenessOneShotRGBEstimation) -> LivenessV1:
"""
Convert a livenessV1 estimation from core result and check error.
Args:
error: error
estimations: core batch livenesv1 estimation
Returns:
livenessv1
"""
assertError(error)
return LivenessV1(estimation,
LivenessPrediction.fromCoreEmotion(estimation.State))
def match(
self,
reference: Union[FaceDescriptor, bytes],
candidates: Union[FaceDescriptor, bytes, List[Union[FaceDescriptor,
bytes]],
FaceDescriptorBatch],
) -> Union[MatchingResult, List[MatchingResult]]:
"""
Match face descriptor vs face descriptors.
Returns:
List of matching results if match by several descriptors otherwise one MatchingResult.
Raises:
TypeError: if candidates has incorrect type
"""
if isinstance(reference, bytes):
referenceForMatcher = self.descriptorFactory.generateDescriptor(
reference)
else:
referenceForMatcher = reference
if isinstance(candidates, bytes):
candidatesForMatcher = self.descriptorFactory.generateDescriptor(
candidates)
error, matchResults = self._coreMatcher.match(
referenceForMatcher.coreEstimation,
candidatesForMatcher.coreEstimation)
elif isinstance(candidates, FaceDescriptor):
error, matchResults = self._coreMatcher.match(
referenceForMatcher.coreEstimation, candidates.coreEstimation)
elif isinstance(candidates, list):
batch = self.descriptorFactory.generateDescriptorsBatch(
len(candidates))
for cand in candidates:
if isinstance(cand, bytes):
candidate = self.descriptorFactory.generateDescriptor(cand)
else:
candidate = cand
batch.append(candidate)
error, matchResults = self._coreMatcher.match(
referenceForMatcher.coreEstimation, batch.coreEstimation)
elif isinstance(candidates, FaceDescriptorBatch):
error, matchResults = self._coreMatcher.match(
referenceForMatcher.coreEstimation, candidates.coreEstimation)
else:
raise TypeError(f"Bad candidates type: {type(candidates)}")
assertError(error)
return matchResults
def estimate(self, warp: Union[FaceWarp, FaceWarpedImage]) -> MouthStates:
"""
Estimate mouth state on warp.
Args:
warp: warped image
Returns:
estimated states
Raises:
LunaSDKException: if estimation failed
"""
error, mouthState = self._coreEstimator.estimate(warp.warpedImage.coreImage)
assertError(error)
return MouthStates(mouthState)
def estimate(self, warp: Union[FaceWarp, FaceWarpedImage]) -> Credibility:
"""
Estimate credibility from a warp.
Args:
warp: raw warped image or warp
Returns:
estimated credibility
Raises:
LunaSDKException: if estimation failed
"""
error, credibility = self._coreEstimator.estimate(warp.warpedImage.coreImage)
assertError(error)
return Credibility(credibility)
def estimateBatch( # type: ignore
self,
faceDetections: List[FaceDetection],
qualityThreshold: Optional[float] = None) -> List[LivenessV1]:
"""
Batch estimate liveness
.. warning::
Current estimator version estimates correct liveness state for images from mobile and web camera only.
A correctness of the liveness prediction is not guaranteed for other image sources.
Args:
faceDetections: face detection list
qualityThreshold: quality threshold. if estimation quality is low of this threshold
Returns:
estimated liveness
Raises:
LunaSDKException: if estimation failed
"""
coreImages = [
detection.image.coreImage for detection in faceDetections
]
detections = [
detection.coreEstimation.detection for detection in faceDetections
]
try:
coreEstimations = [
detection.landmarks5.coreEstimation
for detection in faceDetections
] # type: ignore
except AttributeError:
raise ValueError("Landmarks5 is required for liveness estimation")
validateInputByBatchEstimator(self._coreEstimator, coreImages,
detections, coreEstimations)
error, estimations = self._coreEstimator.estimate(
coreImages,
detections,
coreEstimations,
-1.0 if qualityThreshold is None else qualityThreshold,
)
assertError(error)
return [
LivenessV1(estimation,
LivenessPrediction.fromCoreEmotion(estimation.State))
for estimation in estimations
]
def estimate(self, warp: Union[FaceWarp, FaceWarpedImage]) -> Emotions:
"""
Estimate emotion on warp.
Args:
warp: warped image
Returns:
estimated emotions
Raises:
LunaSDKException: if estimation failed
"""
error, emotions = self._coreEstimator.estimate(
warp.warpedImage.coreImage)
assertError(error)
return Emotions(emotions)
def estimate(self, warpWithLandmarks5: WarpWithLandmarks5) -> GazeDirection:
"""
Estimate a gaze direction
Args:
warpWithLandmarks5: warp with transformed 5 landmarks
Returns:
estimated states
Raises:
LunaSDKException: if estimation failed
"""
error, gaze = self._coreEstimator.estimate(
warpWithLandmarks5.warp.warpedImage.coreImage, warpWithLandmarks5.landmarks.coreEstimation
)
assertError(error)
return GazeDirection(gaze)
def postProcessingBatch(
error: FSDKErrorResult, coreEstimations: List[CoreTaskResult],
resultClass: Type[CommonTaskResult]) -> List[CommonTaskResult]:
"""
Post processing batch estimation
Args:
error: estimation error
coreEstimations: core batch estimation
resultClass: result class
Returns:
list of lunavl structure based on core estimation
"""
assertError(error)
return [resultClass(coreEstimation) for coreEstimation in coreEstimations]
def estimate(self, warp: Union[FaceWarp, FaceWarpedImage]) -> Glasses:
"""
Estimate glasses from a warp.
Args:
warp: raw warped image or warp
Returns:
estimated glasses
Raises:
LunaSDKException: if estimation failed
"""
error, glasses = self._coreEstimator.estimate(
warp.warpedImage.coreImage)
assertError(error)
return Glasses(glasses)
def postProcessingBatch(error: FSDKErrorResult, orientations):
"""
Post processing batch image orientation estimation
Args:
error: estimation error
orientations: estimated orientations
Returns:
list of `OrientationType`
"""
assertError(error)
return [
OrientationType.fromCoreOrientationType(coreOrientationType)
for coreOrientationType in orientations
]
def estimateBasicAttributesBatch(
self,
warps: List[Union[FaceWarp, FaceWarpedImage]],
estimateAge: bool,
estimateGender: bool,
estimateEthnicity: bool,
aggregate: bool = False,
) -> Tuple[List[BasicAttributes], Union[None, BasicAttributes]]:
"""
Batch basic attributes estimation on warped images.
Args:
warps: warped images
estimateAge: estimate age or not
estimateGender: estimate gender or not
estimateEthnicity: estimate ethnicity or not
aggregate: aggregate attributes to one or not
Returns:
tuple, first element - list estimated attributes in corresponding order,
second - optional aggregated attributes.
Raises:
LunaSDKException: if estimation failed
"""
dtAttributes = 0
if estimateAge:
dtAttributes |= AttributeRequest.estimateAge
if estimateGender:
dtAttributes |= AttributeRequest.estimateGender
if estimateEthnicity:
dtAttributes |= AttributeRequest.estimateEthnicity
images = [warp.warpedImage.coreImage for warp in warps]
validateInputByBatchEstimator(self._coreEstimator, images,
AttributeRequest(dtAttributes))
error, baseAttributes, aggregateAttribute = self._coreEstimator.estimate(
images, AttributeRequest(dtAttributes))
assertError(error)
attributes = [
BasicAttributes(baseAttribute) for baseAttribute in baseAttributes
]
if aggregate:
return attributes, BasicAttributes(aggregateAttribute)
else:
return attributes, None
def search(self,
descriptor: FaceDescriptor,
maxCount: int = 1) -> List[IndexResult]:
"""
Search for descriptors with the shorter distance to passed descriptor.
Args:
descriptor: descriptor to match against index
maxCount: max count of results (default is 1)
Raises:
LunaSDKException: if an error occurs while searching for descriptors
Returns:
list with index search results
"""
error, resIndex = self._coreIndex.search(descriptor.coreEstimation,
maxCount)
assertError(error)
return [IndexResult(result) for result in resIndex]