Upload 2 files

cv2/Error/__init__.pyi

@@ -0,0 +1,116 @@
+# Enumerations
+StsOk: int
+STS_OK: int
+StsBackTrace: int
+STS_BACK_TRACE: int
+StsError: int
+STS_ERROR: int
+StsInternal: int
+STS_INTERNAL: int
+StsNoMem: int
+STS_NO_MEM: int
+StsBadArg: int
+STS_BAD_ARG: int
+StsBadFunc: int
+STS_BAD_FUNC: int
+StsNoConv: int
+STS_NO_CONV: int
+StsAutoTrace: int
+STS_AUTO_TRACE: int
+HeaderIsNull: int
+HEADER_IS_NULL: int
+BadImageSize: int
+BAD_IMAGE_SIZE: int
+BadOffset: int
+BAD_OFFSET: int
+BadDataPtr: int
+BAD_DATA_PTR: int
+BadStep: int
+BAD_STEP: int
+BadModelOrChSeq: int
+BAD_MODEL_OR_CH_SEQ: int
+BadNumChannels: int
+BAD_NUM_CHANNELS: int
+BadNumChannel1U: int
+BAD_NUM_CHANNEL1U: int
+BadDepth: int
+BAD_DEPTH: int
+BadAlphaChannel: int
+BAD_ALPHA_CHANNEL: int
+BadOrder: int
+BAD_ORDER: int
+BadOrigin: int
+BAD_ORIGIN: int
+BadAlign: int
+BAD_ALIGN: int
+BadCallBack: int
+BAD_CALL_BACK: int
+BadTileSize: int
+BAD_TILE_SIZE: int
+BadCOI: int
+BAD_COI: int
+BadROISize: int
+BAD_ROISIZE: int
+MaskIsTiled: int
+MASK_IS_TILED: int
+StsNullPtr: int
+STS_NULL_PTR: int
+StsVecLengthErr: int
+STS_VEC_LENGTH_ERR: int
+StsFilterStructContentErr: int
+STS_FILTER_STRUCT_CONTENT_ERR: int
+StsKernelStructContentErr: int
+STS_KERNEL_STRUCT_CONTENT_ERR: int
+StsFilterOffsetErr: int
+STS_FILTER_OFFSET_ERR: int
+StsBadSize: int
+STS_BAD_SIZE: int
+StsDivByZero: int
+STS_DIV_BY_ZERO: int
+StsInplaceNotSupported: int
+STS_INPLACE_NOT_SUPPORTED: int
+StsObjectNotFound: int
+STS_OBJECT_NOT_FOUND: int
+StsUnmatchedFormats: int
+STS_UNMATCHED_FORMATS: int
+StsBadFlag: int
+STS_BAD_FLAG: int
+StsBadPoint: int
+STS_BAD_POINT: int
+StsBadMask: int
+STS_BAD_MASK: int
+StsUnmatchedSizes: int
+STS_UNMATCHED_SIZES: int
+StsUnsupportedFormat: int
+STS_UNSUPPORTED_FORMAT: int
+StsOutOfRange: int
+STS_OUT_OF_RANGE: int
+StsParseError: int
+STS_PARSE_ERROR: int
+StsNotImplemented: int
+STS_NOT_IMPLEMENTED: int
+StsBadMemBlock: int
+STS_BAD_MEM_BLOCK: int
+StsAssert: int
+STS_ASSERT: int
+GpuNotSupported: int
+GPU_NOT_SUPPORTED: int
+GpuApiCallError: int
+GPU_API_CALL_ERROR: int
+OpenGlNotSupported: int
+OPEN_GL_NOT_SUPPORTED: int
+OpenGlApiCallError: int
+OPEN_GL_API_CALL_ERROR: int
+OpenCLApiCallError: int
+OPEN_CLAPI_CALL_ERROR: int
+OpenCLDoubleNotSupported: int
+OPEN_CLDOUBLE_NOT_SUPPORTED: int
+OpenCLInitError: int
+OPEN_CLINIT_ERROR: int
+OpenCLNoAMDBlasFft: int
+OPEN_CLNO_AMDBLAS_FFT: int
+Code = int
+"""One of [StsOk, STS_OK, StsBackTrace, STS_BACK_TRACE, StsError, STS_ERROR, StsInternal, STS_INTERNAL, StsNoMem, STS_NO_MEM, StsBadArg, STS_BAD_ARG, StsBadFunc, STS_BAD_FUNC, StsNoConv, STS_NO_CONV, StsAutoTrace, STS_AUTO_TRACE, HeaderIsNull, HEADER_IS_NULL, BadImageSize, BAD_IMAGE_SIZE, BadOffset, BAD_OFFSET, BadDataPtr, BAD_DATA_PTR, BadStep, BAD_STEP, BadModelOrChSeq, BAD_MODEL_OR_CH_SEQ, BadNumChannels, BAD_NUM_CHANNELS, BadNumChannel1U, BAD_NUM_CHANNEL1U, BadDepth, BAD_DEPTH, BadAlphaChannel, BAD_ALPHA_CHANNEL, BadOrder, BAD_ORDER, BadOrigin, BAD_ORIGIN, BadAlign, BAD_ALIGN, BadCallBack, BAD_CALL_BACK, BadTileSize, BAD_TILE_SIZE, BadCOI, BAD_COI, BadROISize, BAD_ROISIZE, MaskIsTiled, MASK_IS_TILED, StsNullPtr, STS_NULL_PTR, StsVecLengthErr, STS_VEC_LENGTH_ERR, StsFilterStructContentErr, STS_FILTER_STRUCT_CONTENT_ERR, StsKernelStructContentErr, STS_KERNEL_STRUCT_CONTENT_ERR, StsFilterOffsetErr, STS_FILTER_OFFSET_ERR, StsBadSize, STS_BAD_SIZE, StsDivByZero, STS_DIV_BY_ZERO, StsInplaceNotSupported, STS_INPLACE_NOT_SUPPORTED, StsObjectNotFound, STS_OBJECT_NOT_FOUND, StsUnmatchedFormats, STS_UNMATCHED_FORMATS, StsBadFlag, STS_BAD_FLAG, StsBadPoint, STS_BAD_POINT, StsBadMask, STS_BAD_MASK, StsUnmatchedSizes, STS_UNMATCHED_SIZES, StsUnsupportedFormat, STS_UNSUPPORTED_FORMAT, StsOutOfRange, STS_OUT_OF_RANGE, StsParseError, STS_PARSE_ERROR, StsNotImplemented, STS_NOT_IMPLEMENTED, StsBadMemBlock, STS_BAD_MEM_BLOCK, StsAssert, STS_ASSERT, GpuNotSupported, GPU_NOT_SUPPORTED, GpuApiCallError, GPU_API_CALL_ERROR, OpenGlNotSupported, OPEN_GL_NOT_SUPPORTED, OpenGlApiCallError, OPEN_GL_API_CALL_ERROR, OpenCLApiCallError, OPEN_CLAPI_CALL_ERROR, OpenCLDoubleNotSupported, OPEN_CLDOUBLE_NOT_SUPPORTED, OpenCLInitError, OPEN_CLINIT_ERROR, OpenCLNoAMDBlasFft, OPEN_CLNO_AMDBLAS_FFT]"""
+
+
+

cv2/dnn/__init__.pyi

@@ -0,0 +1,503 @@
+import cv2
+import cv2.typing
+import numpy
+import typing
+
+
+# Enumerations
+DNN_BACKEND_DEFAULT: int
+DNN_BACKEND_HALIDE: int
+DNN_BACKEND_INFERENCE_ENGINE: int
+DNN_BACKEND_OPENCV: int
+DNN_BACKEND_VKCOM: int
+DNN_BACKEND_CUDA: int
+DNN_BACKEND_WEBNN: int
+DNN_BACKEND_TIMVX: int
+DNN_BACKEND_CANN: int
+Backend = int
+"""One of [DNN_BACKEND_DEFAULT, DNN_BACKEND_HALIDE, DNN_BACKEND_INFERENCE_ENGINE, DNN_BACKEND_OPENCV, DNN_BACKEND_VKCOM, DNN_BACKEND_CUDA, DNN_BACKEND_WEBNN, DNN_BACKEND_TIMVX, DNN_BACKEND_CANN]"""
+
+DNN_TARGET_CPU: int
+DNN_TARGET_OPENCL: int
+DNN_TARGET_OPENCL_FP16: int
+DNN_TARGET_MYRIAD: int
+DNN_TARGET_VULKAN: int
+DNN_TARGET_FPGA: int
+DNN_TARGET_CUDA: int
+DNN_TARGET_CUDA_FP16: int
+DNN_TARGET_HDDL: int
+DNN_TARGET_NPU: int
+DNN_TARGET_CPU_FP16: int
+Target = int
+"""One of [DNN_TARGET_CPU, DNN_TARGET_OPENCL, DNN_TARGET_OPENCL_FP16, DNN_TARGET_MYRIAD, DNN_TARGET_VULKAN, DNN_TARGET_FPGA, DNN_TARGET_CUDA, DNN_TARGET_CUDA_FP16, DNN_TARGET_HDDL, DNN_TARGET_NPU, DNN_TARGET_CPU_FP16]"""
+
+DNN_LAYOUT_UNKNOWN: int
+DNN_LAYOUT_ND: int
+DNN_LAYOUT_NCHW: int
+DNN_LAYOUT_NCDHW: int
+DNN_LAYOUT_NHWC: int
+DNN_LAYOUT_NDHWC: int
+DNN_LAYOUT_PLANAR: int
+DataLayout = int
+"""One of [DNN_LAYOUT_UNKNOWN, DNN_LAYOUT_ND, DNN_LAYOUT_NCHW, DNN_LAYOUT_NCDHW, DNN_LAYOUT_NHWC, DNN_LAYOUT_NDHWC, DNN_LAYOUT_PLANAR]"""
+
+DNN_PMODE_NULL: int
+DNN_PMODE_CROP_CENTER: int
+DNN_PMODE_LETTERBOX: int
+ImagePaddingMode = int
+"""One of [DNN_PMODE_NULL, DNN_PMODE_CROP_CENTER, DNN_PMODE_LETTERBOX]"""
+
+SoftNMSMethod_SOFTNMS_LINEAR: int
+SOFT_NMSMETHOD_SOFTNMS_LINEAR: int
+SoftNMSMethod_SOFTNMS_GAUSSIAN: int
+SOFT_NMSMETHOD_SOFTNMS_GAUSSIAN: int
+SoftNMSMethod = int
+"""One of [SoftNMSMethod_SOFTNMS_LINEAR, SOFT_NMSMETHOD_SOFTNMS_LINEAR, SoftNMSMethod_SOFTNMS_GAUSSIAN, SOFT_NMSMETHOD_SOFTNMS_GAUSSIAN]"""
+
+
+
+# Classes
+class DictValue:
+    # Functions
+    @typing.overload
+    def __init__(self, i: int) -> None: ...
+    @typing.overload
+    def __init__(self, p: float) -> None: ...
+    @typing.overload
+    def __init__(self, s: str) -> None: ...
+
+    def isInt(self) -> bool: ...
+
+    def isString(self) -> bool: ...
+
+    def isReal(self) -> bool: ...
+
+    def getIntValue(self, idx: int = ...) -> int: ...
+
+    def getRealValue(self, idx: int = ...) -> float: ...
+
+    def getStringValue(self, idx: int = ...) -> str: ...
+
+
+class Net:
+    # Functions
+    def __init__(self) -> None: ...
+
+    @classmethod
+    @typing.overload
+    def readFromModelOptimizer(cls, xml: str, bin: str) -> Net: ...
+    @classmethod
+    @typing.overload
+    def readFromModelOptimizer(cls, bufferModelConfig: numpy.ndarray[typing.Any, numpy.dtype[numpy.uint8]], bufferWeights: numpy.ndarray[typing.Any, numpy.dtype[numpy.uint8]]) -> Net: ...
+
+    def empty(self) -> bool: ...
+
+    def dump(self) -> str: ...
+
+    def dumpToFile(self, path: str) -> None: ...
+
+    def getLayerId(self, layer: str) -> int: ...
+
+    def getLayerNames(self) -> typing.Sequence[str]: ...
+
+    @typing.overload
+    def getLayer(self, layerId: int) -> Layer: ...
+    @typing.overload
+    def getLayer(self, layerName: str) -> Layer: ...
+    @typing.overload
+    def getLayer(self, layerId: cv2.typing.LayerId) -> Layer: ...
+
+    def connect(self, outPin: str, inpPin: str) -> None: ...
+
+    def setInputsNames(self, inputBlobNames: typing.Sequence[str]) -> None: ...
+
+    def setInputShape(self, inputName: str, shape: cv2.typing.MatShape) -> None: ...
+
+    @typing.overload
+    def forward(self, outputName: str = ...) -> cv2.typing.MatLike: ...
+    @typing.overload
+    def forward(self, outputBlobs: typing.Sequence[cv2.typing.MatLike] | None = ..., outputName: str = ...) -> typing.Sequence[cv2.typing.MatLike]: ...
+    @typing.overload
+    def forward(self, outputBlobs: typing.Sequence[cv2.UMat] | None = ..., outputName: str = ...) -> typing.Sequence[cv2.UMat]: ...
+    @typing.overload
+    def forward(self, outBlobNames: typing.Sequence[str], outputBlobs: typing.Sequence[cv2.typing.MatLike] | None = ...) -> typing.Sequence[cv2.typing.MatLike]: ...
+    @typing.overload
+    def forward(self, outBlobNames: typing.Sequence[str], outputBlobs: typing.Sequence[cv2.UMat] | None = ...) -> typing.Sequence[cv2.UMat]: ...
+
+    def forwardAsync(self, outputName: str = ...) -> cv2.AsyncArray: ...
+
+    def forwardAndRetrieve(self, outBlobNames: typing.Sequence[str]) -> typing.Sequence[typing.Sequence[cv2.typing.MatLike]]: ...
+
+    @typing.overload
+    def quantize(self, calibData: typing.Sequence[cv2.typing.MatLike], inputsDtype: int, outputsDtype: int, perChannel: bool = ...) -> Net: ...
+    @typing.overload
+    def quantize(self, calibData: typing.Sequence[cv2.UMat], inputsDtype: int, outputsDtype: int, perChannel: bool = ...) -> Net: ...
+
+    def getInputDetails(self) -> tuple[typing.Sequence[float], typing.Sequence[int]]: ...
+
+    def getOutputDetails(self) -> tuple[typing.Sequence[float], typing.Sequence[int]]: ...
+
+    def setHalideScheduler(self, scheduler: str) -> None: ...
+
+    def setPreferableBackend(self, backendId: int) -> None: ...
+
+    def setPreferableTarget(self, targetId: int) -> None: ...
+
+    @typing.overload
+    def setInput(self, blob: cv2.typing.MatLike, name: str = ..., scalefactor: float = ..., mean: cv2.typing.Scalar = ...) -> None: ...
+    @typing.overload
+    def setInput(self, blob: cv2.UMat, name: str = ..., scalefactor: float = ..., mean: cv2.typing.Scalar = ...) -> None: ...
+
+    @typing.overload
+    def setParam(self, layer: int, numParam: int, blob: cv2.typing.MatLike) -> None: ...
+    @typing.overload
+    def setParam(self, layerName: str, numParam: int, blob: cv2.typing.MatLike) -> None: ...
+
+    @typing.overload
+    def getParam(self, layer: int, numParam: int = ...) -> cv2.typing.MatLike: ...
+    @typing.overload
+    def getParam(self, layerName: str, numParam: int = ...) -> cv2.typing.MatLike: ...
+
+    def getUnconnectedOutLayers(self) -> typing.Sequence[int]: ...
+
+    def getUnconnectedOutLayersNames(self) -> typing.Sequence[str]: ...
+
+    @typing.overload
+    def getLayersShapes(self, netInputShapes: typing.Sequence[cv2.typing.MatShape]) -> tuple[typing.Sequence[int], typing.Sequence[typing.Sequence[cv2.typing.MatShape]], typing.Sequence[typing.Sequence[cv2.typing.MatShape]]]: ...
+    @typing.overload
+    def getLayersShapes(self, netInputShape: cv2.typing.MatShape) -> tuple[typing.Sequence[int], typing.Sequence[typing.Sequence[cv2.typing.MatShape]], typing.Sequence[typing.Sequence[cv2.typing.MatShape]]]: ...
+
+    @typing.overload
+    def getFLOPS(self, netInputShapes: typing.Sequence[cv2.typing.MatShape]) -> int: ...
+    @typing.overload
+    def getFLOPS(self, netInputShape: cv2.typing.MatShape) -> int: ...
+    @typing.overload
+    def getFLOPS(self, layerId: int, netInputShapes: typing.Sequence[cv2.typing.MatShape]) -> int: ...
+    @typing.overload
+    def getFLOPS(self, layerId: int, netInputShape: cv2.typing.MatShape) -> int: ...
+
+    def getLayerTypes(self) -> typing.Sequence[str]: ...
+
+    def getLayersCount(self, layerType: str) -> int: ...
+
+    @typing.overload
+    def getMemoryConsumption(self, netInputShape: cv2.typing.MatShape) -> tuple[int, int]: ...
+    @typing.overload
+    def getMemoryConsumption(self, layerId: int, netInputShapes: typing.Sequence[cv2.typing.MatShape]) -> tuple[int, int]: ...
+    @typing.overload
+    def getMemoryConsumption(self, layerId: int, netInputShape: cv2.typing.MatShape) -> tuple[int, int]: ...
+
+    def enableFusion(self, fusion: bool) -> None: ...
+
+    def enableWinograd(self, useWinograd: bool) -> None: ...
+
+    def getPerfProfile(self) -> tuple[int, typing.Sequence[float]]: ...
+
+
+class Image2BlobParams:
+    scalefactor: cv2.typing.Scalar
+    size: cv2.typing.Size
+    mean: cv2.typing.Scalar
+    swapRB: bool
+    ddepth: int
+    datalayout: DataLayout
+    paddingmode: ImagePaddingMode
+
+    # Functions
+    @typing.overload
+    def __init__(self) -> None: ...
+    @typing.overload
+    def __init__(self, scalefactor: cv2.typing.Scalar, size: cv2.typing.Size = ..., mean: cv2.typing.Scalar = ..., swapRB: bool = ..., ddepth: int = ..., datalayout: DataLayout = ..., mode: ImagePaddingMode = ...) -> None: ...
+
+
+class Model:
+    # Functions
+    @typing.overload
+    def __init__(self, model: str, config: str = ...) -> None: ...
+    @typing.overload
+    def __init__(self, network: Net) -> None: ...
+
+    @typing.overload
+    def setInputSize(self, size: cv2.typing.Size) -> Model: ...
+    @typing.overload
+    def setInputSize(self, width: int, height: int) -> Model: ...
+
+    def setInputMean(self, mean: cv2.typing.Scalar) -> Model: ...
+
+    def setInputScale(self, scale: cv2.typing.Scalar) -> Model: ...
+
+    def setInputCrop(self, crop: bool) -> Model: ...
+
+    def setInputSwapRB(self, swapRB: bool) -> Model: ...
+
+    def setInputParams(self, scale: float = ..., size: cv2.typing.Size = ..., mean: cv2.typing.Scalar = ..., swapRB: bool = ..., crop: bool = ...) -> None: ...
+
+    @typing.overload
+    def predict(self, frame: cv2.typing.MatLike, outs: typing.Sequence[cv2.typing.MatLike] | None = ...) -> typing.Sequence[cv2.typing.MatLike]: ...
+    @typing.overload
+    def predict(self, frame: cv2.UMat, outs: typing.Sequence[cv2.UMat] | None = ...) -> typing.Sequence[cv2.UMat]: ...
+
+    def setPreferableBackend(self, backendId: Backend) -> Model: ...
+
+    def setPreferableTarget(self, targetId: Target) -> Model: ...
+
+
+class Layer(cv2.Algorithm):
+    blobs: typing.Sequence[cv2.typing.MatLike]
+    @property
+    def name(self) -> str: ...
+    @property
+    def type(self) -> str: ...
+    @property
+    def preferableTarget(self) -> int: ...
+
+    # Functions
+    @typing.overload
+    def finalize(self, inputs: typing.Sequence[cv2.typing.MatLike], outputs: typing.Sequence[cv2.typing.MatLike] | None = ...) -> typing.Sequence[cv2.typing.MatLike]: ...
+    @typing.overload
+    def finalize(self, inputs: typing.Sequence[cv2.UMat], outputs: typing.Sequence[cv2.UMat] | None = ...) -> typing.Sequence[cv2.UMat]: ...
+
+    def run(self, inputs: typing.Sequence[cv2.typing.MatLike], internals: typing.Sequence[cv2.typing.MatLike], outputs: typing.Sequence[cv2.typing.MatLike] | None = ...) -> tuple[typing.Sequence[cv2.typing.MatLike], typing.Sequence[cv2.typing.MatLike]]: ...
+
+    def outputNameToIndex(self, outputName: str) -> int: ...
+
+
+class ClassificationModel(Model):
+    # Functions
+    @typing.overload
+    def __init__(self, model: str, config: str = ...) -> None: ...
+    @typing.overload
+    def __init__(self, network: Net) -> None: ...
+
+    def setEnableSoftmaxPostProcessing(self, enable: bool) -> ClassificationModel: ...
+
+    def getEnableSoftmaxPostProcessing(self) -> bool: ...
+
+    @typing.overload
+    def classify(self, frame: cv2.typing.MatLike) -> tuple[int, float]: ...
+    @typing.overload
+    def classify(self, frame: cv2.UMat) -> tuple[int, float]: ...
+
+
+class KeypointsModel(Model):
+    # Functions
+    @typing.overload
+    def __init__(self, model: str, config: str = ...) -> None: ...
+    @typing.overload
+    def __init__(self, network: Net) -> None: ...
+
+    @typing.overload
+    def estimate(self, frame: cv2.typing.MatLike, thresh: float = ...) -> typing.Sequence[cv2.typing.Point2f]: ...
+    @typing.overload
+    def estimate(self, frame: cv2.UMat, thresh: float = ...) -> typing.Sequence[cv2.typing.Point2f]: ...
+
+
+class SegmentationModel(Model):
+    # Functions
+    @typing.overload
+    def __init__(self, model: str, config: str = ...) -> None: ...
+    @typing.overload
+    def __init__(self, network: Net) -> None: ...
+
+    @typing.overload
+    def segment(self, frame: cv2.typing.MatLike, mask: cv2.typing.MatLike | None = ...) -> cv2.typing.MatLike: ...
+    @typing.overload
+    def segment(self, frame: cv2.UMat, mask: cv2.UMat | None = ...) -> cv2.UMat: ...
+
+
+class DetectionModel(Model):
+    # Functions
+    @typing.overload
+    def __init__(self, model: str, config: str = ...) -> None: ...
+    @typing.overload
+    def __init__(self, network: Net) -> None: ...
+
+    def setNmsAcrossClasses(self, value: bool) -> DetectionModel: ...
+
+    def getNmsAcrossClasses(self) -> bool: ...
+
+    @typing.overload
+    def detect(self, frame: cv2.typing.MatLike, confThreshold: float = ..., nmsThreshold: float = ...) -> tuple[typing.Sequence[int], typing.Sequence[float], typing.Sequence[cv2.typing.Rect]]: ...
+    @typing.overload
+    def detect(self, frame: cv2.UMat, confThreshold: float = ..., nmsThreshold: float = ...) -> tuple[typing.Sequence[int], typing.Sequence[float], typing.Sequence[cv2.typing.Rect]]: ...
+
+
+class TextRecognitionModel(Model):
+    # Functions
+    @typing.overload
+    def __init__(self, network: Net) -> None: ...
+    @typing.overload
+    def __init__(self, model: str, config: str = ...) -> None: ...
+
+    def setDecodeType(self, decodeType: str) -> TextRecognitionModel: ...
+
+    def getDecodeType(self) -> str: ...
+
+    def setDecodeOptsCTCPrefixBeamSearch(self, beamSize: int, vocPruneSize: int = ...) -> TextRecognitionModel: ...
+
+    def setVocabulary(self, vocabulary: typing.Sequence[str]) -> TextRecognitionModel: ...
+
+    def getVocabulary(self) -> typing.Sequence[str]: ...
+
+    @typing.overload
+    def recognize(self, frame: cv2.typing.MatLike) -> str: ...
+    @typing.overload
+    def recognize(self, frame: cv2.UMat) -> str: ...
+    @typing.overload
+    def recognize(self, frame: cv2.typing.MatLike, roiRects: typing.Sequence[cv2.typing.MatLike]) -> typing.Sequence[str]: ...
+    @typing.overload
+    def recognize(self, frame: cv2.UMat, roiRects: typing.Sequence[cv2.UMat]) -> typing.Sequence[str]: ...
+
+
+class TextDetectionModel(Model):
+    # Functions
+    @typing.overload
+    def detect(self, frame: cv2.typing.MatLike) -> tuple[typing.Sequence[typing.Sequence[cv2.typing.Point]], typing.Sequence[float]]: ...
+    @typing.overload
+    def detect(self, frame: cv2.UMat) -> tuple[typing.Sequence[typing.Sequence[cv2.typing.Point]], typing.Sequence[float]]: ...
+    @typing.overload
+    def detect(self, frame: cv2.typing.MatLike) -> typing.Sequence[typing.Sequence[cv2.typing.Point]]: ...
+    @typing.overload
+    def detect(self, frame: cv2.UMat) -> typing.Sequence[typing.Sequence[cv2.typing.Point]]: ...
+
+    @typing.overload
+    def detectTextRectangles(self, frame: cv2.typing.MatLike) -> tuple[typing.Sequence[cv2.typing.RotatedRect], typing.Sequence[float]]: ...
+    @typing.overload
+    def detectTextRectangles(self, frame: cv2.UMat) -> tuple[typing.Sequence[cv2.typing.RotatedRect], typing.Sequence[float]]: ...
+    @typing.overload
+    def detectTextRectangles(self, frame: cv2.typing.MatLike) -> typing.Sequence[cv2.typing.RotatedRect]: ...
+    @typing.overload
+    def detectTextRectangles(self, frame: cv2.UMat) -> typing.Sequence[cv2.typing.RotatedRect]: ...
+
+
+class TextDetectionModel_EAST(TextDetectionModel):
+    # Functions
+    @typing.overload
+    def __init__(self, network: Net) -> None: ...
+    @typing.overload
+    def __init__(self, model: str, config: str = ...) -> None: ...
+
+    def setConfidenceThreshold(self, confThreshold: float) -> TextDetectionModel_EAST: ...
+
+    def getConfidenceThreshold(self) -> float: ...
+
+    def setNMSThreshold(self, nmsThreshold: float) -> TextDetectionModel_EAST: ...
+
+    def getNMSThreshold(self) -> float: ...
+
+
+class TextDetectionModel_DB(TextDetectionModel):
+    # Functions
+    @typing.overload
+    def __init__(self, network: Net) -> None: ...
+    @typing.overload
+    def __init__(self, model: str, config: str = ...) -> None: ...
+
+    def setBinaryThreshold(self, binaryThreshold: float) -> TextDetectionModel_DB: ...
+
+    def getBinaryThreshold(self) -> float: ...
+
+    def setPolygonThreshold(self, polygonThreshold: float) -> TextDetectionModel_DB: ...
+
+    def getPolygonThreshold(self) -> float: ...
+
+    def setUnclipRatio(self, unclipRatio: float) -> TextDetectionModel_DB: ...
+
+    def getUnclipRatio(self) -> float: ...
+
+    def setMaxCandidates(self, maxCandidates: int) -> TextDetectionModel_DB: ...
+
+    def getMaxCandidates(self) -> int: ...
+
+
+
+# Functions
+def NMSBoxes(bboxes: typing.Sequence[cv2.typing.Rect2d], scores: typing.Sequence[float], score_threshold: float, nms_threshold: float, eta: float = ..., top_k: int = ...) -> typing.Sequence[int]: ...
+
+def NMSBoxesBatched(bboxes: typing.Sequence[cv2.typing.Rect2d], scores: typing.Sequence[float], class_ids: typing.Sequence[int], score_threshold: float, nms_threshold: float, eta: float = ..., top_k: int = ...) -> typing.Sequence[int]: ...
+
+def NMSBoxesRotated(bboxes: typing.Sequence[cv2.typing.RotatedRect], scores: typing.Sequence[float], score_threshold: float, nms_threshold: float, eta: float = ..., top_k: int = ...) -> typing.Sequence[int]: ...
+
+@typing.overload
+def blobFromImage(image: cv2.typing.MatLike, scalefactor: float = ..., size: cv2.typing.Size = ..., mean: cv2.typing.Scalar = ..., swapRB: bool = ..., crop: bool = ..., ddepth: int = ...) -> cv2.typing.MatLike: ...
+@typing.overload
+def blobFromImage(image: cv2.UMat, scalefactor: float = ..., size: cv2.typing.Size = ..., mean: cv2.typing.Scalar = ..., swapRB: bool = ..., crop: bool = ..., ddepth: int = ...) -> cv2.typing.MatLike: ...
+
+@typing.overload
+def blobFromImageWithParams(image: cv2.typing.MatLike, param: Image2BlobParams = ...) -> cv2.typing.MatLike: ...
+@typing.overload
+def blobFromImageWithParams(image: cv2.UMat, param: Image2BlobParams = ...) -> cv2.typing.MatLike: ...
+@typing.overload
+def blobFromImageWithParams(image: cv2.typing.MatLike, blob: cv2.typing.MatLike | None = ..., param: Image2BlobParams = ...) -> cv2.typing.MatLike: ...
+@typing.overload
+def blobFromImageWithParams(image: cv2.UMat, blob: cv2.UMat | None = ..., param: Image2BlobParams = ...) -> cv2.UMat: ...
+
+@typing.overload
+def blobFromImages(images: typing.Sequence[cv2.typing.MatLike], scalefactor: float = ..., size: cv2.typing.Size = ..., mean: cv2.typing.Scalar = ..., swapRB: bool = ..., crop: bool = ..., ddepth: int = ...) -> cv2.typing.MatLike: ...
+@typing.overload
+def blobFromImages(images: typing.Sequence[cv2.UMat], scalefactor: float = ..., size: cv2.typing.Size = ..., mean: cv2.typing.Scalar = ..., swapRB: bool = ..., crop: bool = ..., ddepth: int = ...) -> cv2.typing.MatLike: ...
+
+@typing.overload
+def blobFromImagesWithParams(images: typing.Sequence[cv2.typing.MatLike], param: Image2BlobParams = ...) -> cv2.typing.MatLike: ...
+@typing.overload
+def blobFromImagesWithParams(images: typing.Sequence[cv2.UMat], param: Image2BlobParams = ...) -> cv2.typing.MatLike: ...
+@typing.overload
+def blobFromImagesWithParams(images: typing.Sequence[cv2.typing.MatLike], blob: cv2.typing.MatLike | None = ..., param: Image2BlobParams = ...) -> cv2.typing.MatLike: ...
+@typing.overload
+def blobFromImagesWithParams(images: typing.Sequence[cv2.UMat], blob: cv2.UMat | None = ..., param: Image2BlobParams = ...) -> cv2.UMat: ...
+
+def getAvailableTargets(be: Backend) -> typing.Sequence[Target]: ...
+
+@typing.overload
+def imagesFromBlob(blob_: cv2.typing.MatLike, images_: typing.Sequence[cv2.typing.MatLike] | None = ...) -> typing.Sequence[cv2.typing.MatLike]: ...
+@typing.overload
+def imagesFromBlob(blob_: cv2.typing.MatLike, images_: typing.Sequence[cv2.UMat] | None = ...) -> typing.Sequence[cv2.UMat]: ...
+
+@typing.overload
+def readNet(model: str, config: str = ..., framework: str = ...) -> Net: ...
+@typing.overload
+def readNet(framework: str, bufferModel: numpy.ndarray[typing.Any, numpy.dtype[numpy.uint8]], bufferConfig: numpy.ndarray[typing.Any, numpy.dtype[numpy.uint8]] = ...) -> Net: ...
+
+@typing.overload
+def readNetFromCaffe(prototxt: str, caffeModel: str = ...) -> Net: ...
+@typing.overload
+def readNetFromCaffe(bufferProto: numpy.ndarray[typing.Any, numpy.dtype[numpy.uint8]], bufferModel: numpy.ndarray[typing.Any, numpy.dtype[numpy.uint8]] = ...) -> Net: ...
+
+@typing.overload
+def readNetFromDarknet(cfgFile: str, darknetModel: str = ...) -> Net: ...
+@typing.overload
+def readNetFromDarknet(bufferCfg: numpy.ndarray[typing.Any, numpy.dtype[numpy.uint8]], bufferModel: numpy.ndarray[typing.Any, numpy.dtype[numpy.uint8]] = ...) -> Net: ...
+
+@typing.overload
+def readNetFromModelOptimizer(xml: str, bin: str) -> Net: ...
+@typing.overload
+def readNetFromModelOptimizer(bufferModelConfig: numpy.ndarray[typing.Any, numpy.dtype[numpy.uint8]], bufferWeights: numpy.ndarray[typing.Any, numpy.dtype[numpy.uint8]]) -> Net: ...
+
+@typing.overload
+def readNetFromONNX(onnxFile: str) -> Net: ...
+@typing.overload
+def readNetFromONNX(buffer: numpy.ndarray[typing.Any, numpy.dtype[numpy.uint8]]) -> Net: ...
+
+@typing.overload
+def readNetFromTFLite(model: str) -> Net: ...
+@typing.overload
+def readNetFromTFLite(bufferModel: numpy.ndarray[typing.Any, numpy.dtype[numpy.uint8]]) -> Net: ...
+
+@typing.overload
+def readNetFromTensorflow(model: str, config: str = ...) -> Net: ...
+@typing.overload
+def readNetFromTensorflow(bufferModel: numpy.ndarray[typing.Any, numpy.dtype[numpy.uint8]], bufferConfig: numpy.ndarray[typing.Any, numpy.dtype[numpy.uint8]] = ...) -> Net: ...
+
+def readNetFromTorch(model: str, isBinary: bool = ..., evaluate: bool = ...) -> Net: ...
+
+def readTensorFromONNX(path: str) -> cv2.typing.MatLike: ...
+
+def readTorchBlob(filename: str, isBinary: bool = ...) -> cv2.typing.MatLike: ...
+
+def shrinkCaffeModel(src: str, dst: str, layersTypes: typing.Sequence[str] = ...) -> None: ...
+
+def softNMSBoxes(bboxes: typing.Sequence[cv2.typing.Rect], scores: typing.Sequence[float], score_threshold: float, nms_threshold: float, top_k: int = ..., sigma: float = ..., method: SoftNMSMethod = ...) -> tuple[typing.Sequence[float], typing.Sequence[int]]: ...
+
+def writeTextGraph(model: str, output: str) -> None: ...
+
+