opencv-contrib-python 4.12.0.88__cp37-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl

cv2/face/__init__.pyi

@@ -0,0 +1,219 @@
+__all__: list[str] = []
+
+import cv2
+import cv2.typing
+import typing as _typing
+
+
+# Classes
+class FaceRecognizer(cv2.Algorithm):
+    # Functions
+    @_typing.overload
+    def train(self, src: _typing.Sequence[cv2.typing.MatLike], labels: cv2.typing.MatLike) -> None: ...
+    @_typing.overload
+    def train(self, src: _typing.Sequence[cv2.UMat], labels: cv2.UMat) -> None: ...
+
+    @_typing.overload
+    def update(self, src: _typing.Sequence[cv2.typing.MatLike], labels: cv2.typing.MatLike) -> None: ...
+    @_typing.overload
+    def update(self, src: _typing.Sequence[cv2.UMat], labels: cv2.UMat) -> None: ...
+
+    @_typing.overload
+    def predict_label(self, src: cv2.typing.MatLike) -> int: ...
+    @_typing.overload
+    def predict_label(self, src: cv2.UMat) -> int: ...
+
+    @_typing.overload
+    def predict(self, src: cv2.typing.MatLike) -> tuple[int, float]: ...
+    @_typing.overload
+    def predict(self, src: cv2.UMat) -> tuple[int, float]: ...
+
+    @_typing.overload
+    def predict_collect(self, src: cv2.typing.MatLike, collector: PredictCollector) -> None: ...
+    @_typing.overload
+    def predict_collect(self, src: cv2.UMat, collector: PredictCollector) -> None: ...
+
+    def write(self, filename: str) -> None: ...
+
+    def read(self, filename: str) -> None: ...
+
+    def setLabelInfo(self, label: int, strInfo: str) -> None: ...
+
+    def getLabelInfo(self, label: int) -> str: ...
+
+    def getLabelsByString(self, str: str) -> _typing.Sequence[int]: ...
+
+
+class BIF(cv2.Algorithm):
+    # Functions
+    def getNumBands(self) -> int: ...
+
+    def getNumRotations(self) -> int: ...
+
+    @_typing.overload
+    def compute(self, image: cv2.typing.MatLike, features: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+    @_typing.overload
+    def compute(self, image: cv2.UMat, features: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+    @classmethod
+    def create(cls, num_bands: int = ..., num_rotations: int = ...) -> BIF: ...
+
+
+class FacemarkKazemi(Facemark):
+    ...
+
+class Facemark(cv2.Algorithm):
+    # Functions
+    def loadModel(self, model: str) -> None: ...
+
+    @_typing.overload
+    def fit(self, image: cv2.typing.MatLike, faces: cv2.typing.MatLike, landmarks: _typing.Sequence[cv2.typing.MatLike] | None = ...) -> tuple[bool, _typing.Sequence[cv2.typing.MatLike]]: ...
+    @_typing.overload
+    def fit(self, image: cv2.UMat, faces: cv2.UMat, landmarks: _typing.Sequence[cv2.UMat] | None = ...) -> tuple[bool, _typing.Sequence[cv2.UMat]]: ...
+
+
+class FacemarkAAM(FacemarkTrain):
+    ...
+
+class FacemarkTrain(Facemark):
+    ...
+
+class FacemarkLBF(FacemarkTrain):
+    ...
+
+class BasicFaceRecognizer(FaceRecognizer):
+    # Functions
+    def getNumComponents(self) -> int: ...
+
+    def setNumComponents(self, val: int) -> None: ...
+
+    def getThreshold(self) -> float: ...
+
+    def setThreshold(self, val: float) -> None: ...
+
+    def getProjections(self) -> _typing.Sequence[cv2.typing.MatLike]: ...
+
+    def getLabels(self) -> cv2.typing.MatLike: ...
+
+    def getEigenValues(self) -> cv2.typing.MatLike: ...
+
+    def getEigenVectors(self) -> cv2.typing.MatLike: ...
+
+    def getMean(self) -> cv2.typing.MatLike: ...
+
+
+class EigenFaceRecognizer(BasicFaceRecognizer):
+    # Functions
+    @classmethod
+    def create(cls, num_components: int = ..., threshold: float = ...) -> EigenFaceRecognizer: ...
+
+
+class FisherFaceRecognizer(BasicFaceRecognizer):
+    # Functions
+    @classmethod
+    def create(cls, num_components: int = ..., threshold: float = ...) -> FisherFaceRecognizer: ...
+
+
+class LBPHFaceRecognizer(FaceRecognizer):
+    # Functions
+    def getGridX(self) -> int: ...
+
+    def setGridX(self, val: int) -> None: ...
+
+    def getGridY(self) -> int: ...
+
+    def setGridY(self, val: int) -> None: ...
+
+    def getRadius(self) -> int: ...
+
+    def setRadius(self, val: int) -> None: ...
+
+    def getNeighbors(self) -> int: ...
+
+    def setNeighbors(self, val: int) -> None: ...
+
+    def getThreshold(self) -> float: ...
+
+    def setThreshold(self, val: float) -> None: ...
+
+    def getHistograms(self) -> _typing.Sequence[cv2.typing.MatLike]: ...
+
+    def getLabels(self) -> cv2.typing.MatLike: ...
+
+    @classmethod
+    def create(cls, radius: int = ..., neighbors: int = ..., grid_x: int = ..., grid_y: int = ..., threshold: float = ...) -> LBPHFaceRecognizer: ...
+
+
+class MACE(cv2.Algorithm):
+    # Functions
+    def salt(self, passphrase: str) -> None: ...
+
+    @_typing.overload
+    def train(self, images: _typing.Sequence[cv2.typing.MatLike]) -> None: ...
+    @_typing.overload
+    def train(self, images: _typing.Sequence[cv2.UMat]) -> None: ...
+
+    @_typing.overload
+    def same(self, query: cv2.typing.MatLike) -> bool: ...
+    @_typing.overload
+    def same(self, query: cv2.UMat) -> bool: ...
+
+    @classmethod
+    def load(cls, filename: str, objname: str = ...) -> MACE: ...
+
+    @classmethod
+    def create(cls, IMGSIZE: int = ...) -> MACE: ...
+
+
+class PredictCollector:
+    ...
+
+class StandardCollector(PredictCollector):
+    # Functions
+    def getMinLabel(self) -> int: ...
+
+    def getMinDist(self) -> float: ...
+
+    def getResults(self, sorted: bool = ...) -> _typing.Sequence[tuple[int, float]]: ...
+
+    @classmethod
+    def create(cls, threshold: float = ...) -> StandardCollector: ...
+
+
+
+# Functions
+def createFacemarkAAM() -> Facemark: ...
+
+def createFacemarkKazemi() -> Facemark: ...
+
+def createFacemarkLBF() -> Facemark: ...
+
+@_typing.overload
+def drawFacemarks(image: cv2.typing.MatLike, points: cv2.typing.MatLike, color: cv2.typing.Scalar = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def drawFacemarks(image: cv2.UMat, points: cv2.UMat, color: cv2.typing.Scalar = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def getFacesHAAR(image: cv2.typing.MatLike, face_cascade_name: str, faces: cv2.typing.MatLike | None = ...) -> tuple[bool, cv2.typing.MatLike]: ...
+@_typing.overload
+def getFacesHAAR(image: cv2.UMat, face_cascade_name: str, faces: cv2.UMat | None = ...) -> tuple[bool, cv2.UMat]: ...
+
+def loadDatasetList(imageList: str, annotationList: str, images: _typing.Sequence[str], annotations: _typing.Sequence[str]) -> bool: ...
+
+@_typing.overload
+def loadFacePoints(filename: str, points: cv2.typing.MatLike | None = ..., offset: float = ...) -> tuple[bool, cv2.typing.MatLike]: ...
+@_typing.overload
+def loadFacePoints(filename: str, points: cv2.UMat | None = ..., offset: float = ...) -> tuple[bool, cv2.UMat]: ...
+
+@_typing.overload
+def loadTrainingData(filename: str, images: _typing.Sequence[str], facePoints: cv2.typing.MatLike | None = ..., delim: str = ..., offset: float = ...) -> tuple[bool, cv2.typing.MatLike]: ...
+@_typing.overload
+def loadTrainingData(filename: str, images: _typing.Sequence[str], facePoints: cv2.UMat | None = ..., delim: str = ..., offset: float = ...) -> tuple[bool, cv2.UMat]: ...
+@_typing.overload
+def loadTrainingData(imageList: str, groundTruth: str, images: _typing.Sequence[str], facePoints: cv2.typing.MatLike | None = ..., offset: float = ...) -> tuple[bool, cv2.typing.MatLike]: ...
+@_typing.overload
+def loadTrainingData(imageList: str, groundTruth: str, images: _typing.Sequence[str], facePoints: cv2.UMat | None = ..., offset: float = ...) -> tuple[bool, cv2.UMat]: ...
+@_typing.overload
+def loadTrainingData(filename: _typing.Sequence[str], trainlandmarks: _typing.Sequence[_typing.Sequence[cv2.typing.Point2f]], trainimages: _typing.Sequence[str]) -> bool: ...
+
+

cv2/fisheye/__init__.pyi

@@ -0,0 +1,88 @@
+__all__: list[str] = []
+
+import cv2
+import cv2.typing
+import typing as _typing
+
+
+# Enumerations
+CALIB_USE_INTRINSIC_GUESS: int
+CALIB_RECOMPUTE_EXTRINSIC: int
+CALIB_CHECK_COND: int
+CALIB_FIX_SKEW: int
+CALIB_FIX_K1: int
+CALIB_FIX_K2: int
+CALIB_FIX_K3: int
+CALIB_FIX_K4: int
+CALIB_FIX_INTRINSIC: int
+CALIB_FIX_PRINCIPAL_POINT: int
+CALIB_ZERO_DISPARITY: int
+CALIB_FIX_FOCAL_LENGTH: int
+
+
+
+# Functions
+@_typing.overload
+def calibrate(objectPoints: _typing.Sequence[cv2.typing.MatLike], imagePoints: _typing.Sequence[cv2.typing.MatLike], image_size: cv2.typing.Size, K: cv2.typing.MatLike, D: cv2.typing.MatLike, rvecs: _typing.Sequence[cv2.typing.MatLike] | None = ..., tvecs: _typing.Sequence[cv2.typing.MatLike] | None = ..., flags: int = ..., criteria: cv2.typing.TermCriteria = ...) -> tuple[float, cv2.typing.MatLike, cv2.typing.MatLike, _typing.Sequence[cv2.typing.MatLike], _typing.Sequence[cv2.typing.MatLike]]: ...
+@_typing.overload
+def calibrate(objectPoints: _typing.Sequence[cv2.UMat], imagePoints: _typing.Sequence[cv2.UMat], image_size: cv2.typing.Size, K: cv2.UMat, D: cv2.UMat, rvecs: _typing.Sequence[cv2.UMat] | None = ..., tvecs: _typing.Sequence[cv2.UMat] | None = ..., flags: int = ..., criteria: cv2.typing.TermCriteria = ...) -> tuple[float, cv2.UMat, cv2.UMat, _typing.Sequence[cv2.UMat], _typing.Sequence[cv2.UMat]]: ...
+
+@_typing.overload
+def distortPoints(undistorted: cv2.typing.MatLike, K: cv2.typing.MatLike, D: cv2.typing.MatLike, distorted: cv2.typing.MatLike | None = ..., alpha: float = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def distortPoints(undistorted: cv2.UMat, K: cv2.UMat, D: cv2.UMat, distorted: cv2.UMat | None = ..., alpha: float = ...) -> cv2.UMat: ...
+@_typing.overload
+def distortPoints(undistorted: cv2.typing.MatLike, Kundistorted: cv2.typing.MatLike, K: cv2.typing.MatLike, D: cv2.typing.MatLike, distorted: cv2.typing.MatLike | None = ..., alpha: float = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def distortPoints(undistorted: cv2.UMat, Kundistorted: cv2.UMat, K: cv2.UMat, D: cv2.UMat, distorted: cv2.UMat | None = ..., alpha: float = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def estimateNewCameraMatrixForUndistortRectify(K: cv2.typing.MatLike, D: cv2.typing.MatLike, image_size: cv2.typing.Size, R: cv2.typing.MatLike, P: cv2.typing.MatLike | None = ..., balance: float = ..., new_size: cv2.typing.Size = ..., fov_scale: float = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def estimateNewCameraMatrixForUndistortRectify(K: cv2.UMat, D: cv2.UMat, image_size: cv2.typing.Size, R: cv2.UMat, P: cv2.UMat | None = ..., balance: float = ..., new_size: cv2.typing.Size = ..., fov_scale: float = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def initUndistortRectifyMap(K: cv2.typing.MatLike, D: cv2.typing.MatLike, R: cv2.typing.MatLike, P: cv2.typing.MatLike, size: cv2.typing.Size, m1type: int, map1: cv2.typing.MatLike | None = ..., map2: cv2.typing.MatLike | None = ...) -> tuple[cv2.typing.MatLike, cv2.typing.MatLike]: ...
+@_typing.overload
+def initUndistortRectifyMap(K: cv2.UMat, D: cv2.UMat, R: cv2.UMat, P: cv2.UMat, size: cv2.typing.Size, m1type: int, map1: cv2.UMat | None = ..., map2: cv2.UMat | None = ...) -> tuple[cv2.UMat, cv2.UMat]: ...
+
+@_typing.overload
+def projectPoints(objectPoints: cv2.typing.MatLike, rvec: cv2.typing.MatLike, tvec: cv2.typing.MatLike, K: cv2.typing.MatLike, D: cv2.typing.MatLike, imagePoints: cv2.typing.MatLike | None = ..., alpha: float = ..., jacobian: cv2.typing.MatLike | None = ...) -> tuple[cv2.typing.MatLike, cv2.typing.MatLike]: ...
+@_typing.overload
+def projectPoints(objectPoints: cv2.UMat, rvec: cv2.UMat, tvec: cv2.UMat, K: cv2.UMat, D: cv2.UMat, imagePoints: cv2.UMat | None = ..., alpha: float = ..., jacobian: cv2.UMat | None = ...) -> tuple[cv2.UMat, cv2.UMat]: ...
+
+@_typing.overload
+def solvePnP(objectPoints: cv2.typing.MatLike, imagePoints: cv2.typing.MatLike, cameraMatrix: cv2.typing.MatLike, distCoeffs: cv2.typing.MatLike, rvec: cv2.typing.MatLike | None = ..., tvec: cv2.typing.MatLike | None = ..., useExtrinsicGuess: bool = ..., flags: int = ..., criteria: cv2.typing.TermCriteria = ...) -> tuple[bool, cv2.typing.MatLike, cv2.typing.MatLike]: ...
+@_typing.overload
+def solvePnP(objectPoints: cv2.UMat, imagePoints: cv2.UMat, cameraMatrix: cv2.UMat, distCoeffs: cv2.UMat, rvec: cv2.UMat | None = ..., tvec: cv2.UMat | None = ..., useExtrinsicGuess: bool = ..., flags: int = ..., criteria: cv2.typing.TermCriteria = ...) -> tuple[bool, cv2.UMat, cv2.UMat]: ...
+
+@_typing.overload
+def solvePnPRansac(objectPoints: cv2.typing.MatLike, imagePoints: cv2.typing.MatLike, cameraMatrix: cv2.typing.MatLike, distCoeffs: cv2.typing.MatLike, rvec: cv2.typing.MatLike | None = ..., tvec: cv2.typing.MatLike | None = ..., useExtrinsicGuess: bool = ..., iterationsCount: int = ..., reprojectionError: float = ..., confidence: float = ..., inliers: cv2.typing.MatLike | None = ..., flags: int = ..., criteria: cv2.typing.TermCriteria = ...) -> tuple[bool, cv2.typing.MatLike, cv2.typing.MatLike, cv2.typing.MatLike]: ...
+@_typing.overload
+def solvePnPRansac(objectPoints: cv2.UMat, imagePoints: cv2.UMat, cameraMatrix: cv2.UMat, distCoeffs: cv2.UMat, rvec: cv2.UMat | None = ..., tvec: cv2.UMat | None = ..., useExtrinsicGuess: bool = ..., iterationsCount: int = ..., reprojectionError: float = ..., confidence: float = ..., inliers: cv2.UMat | None = ..., flags: int = ..., criteria: cv2.typing.TermCriteria = ...) -> tuple[bool, cv2.UMat, cv2.UMat, cv2.UMat]: ...
+
+@_typing.overload
+def stereoCalibrate(objectPoints: _typing.Sequence[cv2.typing.MatLike], imagePoints1: _typing.Sequence[cv2.typing.MatLike], imagePoints2: _typing.Sequence[cv2.typing.MatLike], K1: cv2.typing.MatLike, D1: cv2.typing.MatLike, K2: cv2.typing.MatLike, D2: cv2.typing.MatLike, imageSize: cv2.typing.Size, R: cv2.typing.MatLike | None = ..., T: cv2.typing.MatLike | None = ..., rvecs: _typing.Sequence[cv2.typing.MatLike] | None = ..., tvecs: _typing.Sequence[cv2.typing.MatLike] | None = ..., flags: int = ..., criteria: cv2.typing.TermCriteria = ...) -> tuple[float, cv2.typing.MatLike, cv2.typing.MatLike, cv2.typing.MatLike, cv2.typing.MatLike, cv2.typing.MatLike, cv2.typing.MatLike, _typing.Sequence[cv2.typing.MatLike], _typing.Sequence[cv2.typing.MatLike]]: ...
+@_typing.overload
+def stereoCalibrate(objectPoints: _typing.Sequence[cv2.UMat], imagePoints1: _typing.Sequence[cv2.UMat], imagePoints2: _typing.Sequence[cv2.UMat], K1: cv2.UMat, D1: cv2.UMat, K2: cv2.UMat, D2: cv2.UMat, imageSize: cv2.typing.Size, R: cv2.UMat | None = ..., T: cv2.UMat | None = ..., rvecs: _typing.Sequence[cv2.UMat] | None = ..., tvecs: _typing.Sequence[cv2.UMat] | None = ..., flags: int = ..., criteria: cv2.typing.TermCriteria = ...) -> tuple[float, cv2.UMat, cv2.UMat, cv2.UMat, cv2.UMat, cv2.UMat, cv2.UMat, _typing.Sequence[cv2.UMat], _typing.Sequence[cv2.UMat]]: ...
+@_typing.overload
+def stereoCalibrate(objectPoints: _typing.Sequence[cv2.typing.MatLike], imagePoints1: _typing.Sequence[cv2.typing.MatLike], imagePoints2: _typing.Sequence[cv2.typing.MatLike], K1: cv2.typing.MatLike, D1: cv2.typing.MatLike, K2: cv2.typing.MatLike, D2: cv2.typing.MatLike, imageSize: cv2.typing.Size, R: cv2.typing.MatLike | None = ..., T: cv2.typing.MatLike | None = ..., flags: int = ..., criteria: cv2.typing.TermCriteria = ...) -> tuple[float, cv2.typing.MatLike, cv2.typing.MatLike, cv2.typing.MatLike, cv2.typing.MatLike, cv2.typing.MatLike, cv2.typing.MatLike]: ...
+@_typing.overload
+def stereoCalibrate(objectPoints: _typing.Sequence[cv2.UMat], imagePoints1: _typing.Sequence[cv2.UMat], imagePoints2: _typing.Sequence[cv2.UMat], K1: cv2.UMat, D1: cv2.UMat, K2: cv2.UMat, D2: cv2.UMat, imageSize: cv2.typing.Size, R: cv2.UMat | None = ..., T: cv2.UMat | None = ..., flags: int = ..., criteria: cv2.typing.TermCriteria = ...) -> tuple[float, cv2.UMat, cv2.UMat, cv2.UMat, cv2.UMat, cv2.UMat, cv2.UMat]: ...
+
+@_typing.overload
+def stereoRectify(K1: cv2.typing.MatLike, D1: cv2.typing.MatLike, K2: cv2.typing.MatLike, D2: cv2.typing.MatLike, imageSize: cv2.typing.Size, R: cv2.typing.MatLike, tvec: cv2.typing.MatLike, flags: int, R1: cv2.typing.MatLike | None = ..., R2: cv2.typing.MatLike | None = ..., P1: cv2.typing.MatLike | None = ..., P2: cv2.typing.MatLike | None = ..., Q: cv2.typing.MatLike | None = ..., newImageSize: cv2.typing.Size = ..., balance: float = ..., fov_scale: float = ...) -> tuple[cv2.typing.MatLike, cv2.typing.MatLike, cv2.typing.MatLike, cv2.typing.MatLike, cv2.typing.MatLike]: ...
+@_typing.overload
+def stereoRectify(K1: cv2.UMat, D1: cv2.UMat, K2: cv2.UMat, D2: cv2.UMat, imageSize: cv2.typing.Size, R: cv2.UMat, tvec: cv2.UMat, flags: int, R1: cv2.UMat | None = ..., R2: cv2.UMat | None = ..., P1: cv2.UMat | None = ..., P2: cv2.UMat | None = ..., Q: cv2.UMat | None = ..., newImageSize: cv2.typing.Size = ..., balance: float = ..., fov_scale: float = ...) -> tuple[cv2.UMat, cv2.UMat, cv2.UMat, cv2.UMat, cv2.UMat]: ...
+
+@_typing.overload
+def undistortImage(distorted: cv2.typing.MatLike, K: cv2.typing.MatLike, D: cv2.typing.MatLike, undistorted: cv2.typing.MatLike | None = ..., Knew: cv2.typing.MatLike | None = ..., new_size: cv2.typing.Size = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def undistortImage(distorted: cv2.UMat, K: cv2.UMat, D: cv2.UMat, undistorted: cv2.UMat | None = ..., Knew: cv2.UMat | None = ..., new_size: cv2.typing.Size = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def undistortPoints(distorted: cv2.typing.MatLike, K: cv2.typing.MatLike, D: cv2.typing.MatLike, undistorted: cv2.typing.MatLike | None = ..., R: cv2.typing.MatLike | None = ..., P: cv2.typing.MatLike | None = ..., criteria: cv2.typing.TermCriteria = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def undistortPoints(distorted: cv2.UMat, K: cv2.UMat, D: cv2.UMat, undistorted: cv2.UMat | None = ..., R: cv2.UMat | None = ..., P: cv2.UMat | None = ..., criteria: cv2.typing.TermCriteria = ...) -> cv2.UMat: ...
+
+

cv2/flann/__init__.pyi

@@ -0,0 +1,64 @@
+__all__: list[str] = []
+
+import cv2
+import cv2.typing
+import typing as _typing
+
+
+# Enumerations
+FLANN_INDEX_TYPE_8U: int
+FLANN_INDEX_TYPE_8S: int
+FLANN_INDEX_TYPE_16U: int
+FLANN_INDEX_TYPE_16S: int
+FLANN_INDEX_TYPE_32S: int
+FLANN_INDEX_TYPE_32F: int
+FLANN_INDEX_TYPE_64F: int
+FLANN_INDEX_TYPE_STRING: int
+FLANN_INDEX_TYPE_BOOL: int
+FLANN_INDEX_TYPE_ALGORITHM: int
+LAST_VALUE_FLANN_INDEX_TYPE: int
+FlannIndexType = int
+"""One of [FLANN_INDEX_TYPE_8U, FLANN_INDEX_TYPE_8S, FLANN_INDEX_TYPE_16U, FLANN_INDEX_TYPE_16S, FLANN_INDEX_TYPE_32S, FLANN_INDEX_TYPE_32F, FLANN_INDEX_TYPE_64F, FLANN_INDEX_TYPE_STRING, FLANN_INDEX_TYPE_BOOL, FLANN_INDEX_TYPE_ALGORITHM, LAST_VALUE_FLANN_INDEX_TYPE]"""
+
+
+
+# Classes
+class Index:
+    # Functions
+    @_typing.overload
+    def __init__(self) -> None: ...
+    @_typing.overload
+    def __init__(self, features: cv2.typing.MatLike, params: cv2.typing.IndexParams, distType: int = ...) -> None: ...
+    @_typing.overload
+    def __init__(self, features: cv2.UMat, params: cv2.typing.IndexParams, distType: int = ...) -> None: ...
+
+    @_typing.overload
+    def build(self, features: cv2.typing.MatLike, params: cv2.typing.IndexParams, distType: int = ...) -> None: ...
+    @_typing.overload
+    def build(self, features: cv2.UMat, params: cv2.typing.IndexParams, distType: int = ...) -> None: ...
+
+    @_typing.overload
+    def knnSearch(self, query: cv2.typing.MatLike, knn: int, indices: cv2.typing.MatLike | None = ..., dists: cv2.typing.MatLike | None = ..., params: cv2.typing.SearchParams = ...) -> tuple[cv2.typing.MatLike, cv2.typing.MatLike]: ...
+    @_typing.overload
+    def knnSearch(self, query: cv2.UMat, knn: int, indices: cv2.UMat | None = ..., dists: cv2.UMat | None = ..., params: cv2.typing.SearchParams = ...) -> tuple[cv2.UMat, cv2.UMat]: ...
+
+    @_typing.overload
+    def radiusSearch(self, query: cv2.typing.MatLike, radius: float, maxResults: int, indices: cv2.typing.MatLike | None = ..., dists: cv2.typing.MatLike | None = ..., params: cv2.typing.SearchParams = ...) -> tuple[int, cv2.typing.MatLike, cv2.typing.MatLike]: ...
+    @_typing.overload
+    def radiusSearch(self, query: cv2.UMat, radius: float, maxResults: int, indices: cv2.UMat | None = ..., dists: cv2.UMat | None = ..., params: cv2.typing.SearchParams = ...) -> tuple[int, cv2.UMat, cv2.UMat]: ...
+
+    def save(self, filename: str) -> None: ...
+
+    @_typing.overload
+    def load(self, features: cv2.typing.MatLike, filename: str) -> bool: ...
+    @_typing.overload
+    def load(self, features: cv2.UMat, filename: str) -> bool: ...
+
+    def release(self) -> None: ...
+
+    def getDistance(self) -> int: ...
+
+    def getAlgorithm(self) -> int: ...
+
+
+

cv2/ft/__init__.pyi

@@ -0,0 +1,98 @@
+__all__: list[str] = []
+
+import cv2
+import cv2.typing
+import typing as _typing
+
+
+# Enumerations
+LINEAR: int
+SINUS: int
+ONE_STEP: int
+MULTI_STEP: int
+ITERATIVE: int
+
+
+
+# Functions
+@_typing.overload
+def FT02D_FL_process(matrix: cv2.typing.MatLike, radius: int, output: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def FT02D_FL_process(matrix: cv2.UMat, radius: int, output: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def FT02D_FL_process_float(matrix: cv2.typing.MatLike, radius: int, output: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def FT02D_FL_process_float(matrix: cv2.UMat, radius: int, output: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def FT02D_components(matrix: cv2.typing.MatLike, kernel: cv2.typing.MatLike, components: cv2.typing.MatLike | None = ..., mask: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def FT02D_components(matrix: cv2.UMat, kernel: cv2.UMat, components: cv2.UMat | None = ..., mask: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def FT02D_inverseFT(components: cv2.typing.MatLike, kernel: cv2.typing.MatLike, width: int, height: int, output: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def FT02D_inverseFT(components: cv2.UMat, kernel: cv2.UMat, width: int, height: int, output: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def FT02D_iteration(matrix: cv2.typing.MatLike, kernel: cv2.typing.MatLike, mask: cv2.typing.MatLike, firstStop: bool, output: cv2.typing.MatLike | None = ..., maskOutput: cv2.typing.MatLike | None = ...) -> tuple[int, cv2.typing.MatLike, cv2.typing.MatLike]: ...
+@_typing.overload
+def FT02D_iteration(matrix: cv2.UMat, kernel: cv2.UMat, mask: cv2.UMat, firstStop: bool, output: cv2.UMat | None = ..., maskOutput: cv2.UMat | None = ...) -> tuple[int, cv2.UMat, cv2.UMat]: ...
+
+@_typing.overload
+def FT02D_process(matrix: cv2.typing.MatLike, kernel: cv2.typing.MatLike, output: cv2.typing.MatLike | None = ..., mask: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def FT02D_process(matrix: cv2.UMat, kernel: cv2.UMat, output: cv2.UMat | None = ..., mask: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def FT12D_components(matrix: cv2.typing.MatLike, kernel: cv2.typing.MatLike, components: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def FT12D_components(matrix: cv2.UMat, kernel: cv2.UMat, components: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def FT12D_createPolynomMatrixHorizontal(radius: int, chn: int, matrix: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def FT12D_createPolynomMatrixHorizontal(radius: int, chn: int, matrix: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def FT12D_createPolynomMatrixVertical(radius: int, chn: int, matrix: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def FT12D_createPolynomMatrixVertical(radius: int, chn: int, matrix: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def FT12D_inverseFT(components: cv2.typing.MatLike, kernel: cv2.typing.MatLike, width: int, height: int, output: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def FT12D_inverseFT(components: cv2.UMat, kernel: cv2.UMat, width: int, height: int, output: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def FT12D_polynomial(matrix: cv2.typing.MatLike, kernel: cv2.typing.MatLike, c00: cv2.typing.MatLike | None = ..., c10: cv2.typing.MatLike | None = ..., c01: cv2.typing.MatLike | None = ..., components: cv2.typing.MatLike | None = ..., mask: cv2.typing.MatLike | None = ...) -> tuple[cv2.typing.MatLike, cv2.typing.MatLike, cv2.typing.MatLike, cv2.typing.MatLike]: ...
+@_typing.overload
+def FT12D_polynomial(matrix: cv2.UMat, kernel: cv2.UMat, c00: cv2.UMat | None = ..., c10: cv2.UMat | None = ..., c01: cv2.UMat | None = ..., components: cv2.UMat | None = ..., mask: cv2.UMat | None = ...) -> tuple[cv2.UMat, cv2.UMat, cv2.UMat, cv2.UMat]: ...
+
+@_typing.overload
+def FT12D_process(matrix: cv2.typing.MatLike, kernel: cv2.typing.MatLike, output: cv2.typing.MatLike | None = ..., mask: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def FT12D_process(matrix: cv2.UMat, kernel: cv2.UMat, output: cv2.UMat | None = ..., mask: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def createKernel(function: int, radius: int, chn: int, kernel: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def createKernel(function: int, radius: int, chn: int, kernel: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def createKernel1(A: cv2.typing.MatLike, B: cv2.typing.MatLike, chn: int, kernel: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def createKernel1(A: cv2.UMat, B: cv2.UMat, chn: int, kernel: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def filter(image: cv2.typing.MatLike, kernel: cv2.typing.MatLike, output: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def filter(image: cv2.UMat, kernel: cv2.UMat, output: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+@_typing.overload
+def inpaint(image: cv2.typing.MatLike, mask: cv2.typing.MatLike, radius: int, function: int, algorithm: int, output: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+@_typing.overload
+def inpaint(image: cv2.UMat, mask: cv2.UMat, radius: int, function: int, algorithm: int, output: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+