cv::dnn Namespace Reference

Namespaces
namespace	accessor
namespace	details

Classes
struct	_Range
class	AbsLayer
class	AccumLayer
class	AcoshLayer
class	AcosLayer
class	ActivationLayer
class	ActivationLayerInt8
class	ArgLayer
	ArgMax/ArgMin layer. More...
class	AsinhLayer
class	AsinLayer
class	AtanhLayer
class	AtanLayer
class	AttentionLayer
class	BackendNode
	Derivatives of this class encapsulates functions of certain backends. More...
class	BackendWrapper
	Derivatives of this class wraps cv::Mat for different backends and targets. More...
class	BaseConvolutionLayer
class	BatchNormLayer
class	BatchNormLayerInt8
class	BlankLayer
class	BNLLLayer
class	CeilLayer
class	CeluLayer
class	ChannelsPReLULayer
class	ClassificationModel
	This class represents high-level API for classification models. More...
class	CompareLayer
class	ConcatLayer
class	ConstLayer
class	ConvolutionLayer
class	ConvolutionLayerInt8
class	CorrelationLayer
class	CoshLayer
class	CosLayer
class	CropAndResizeLayer
class	CropLayer
class	CumSumLayer
class	DataAugmentationLayer
class	DeconvolutionLayer
class	DequantizeLayer
class	DetectionModel
	This class represents high-level API for object detection networks. More...
class	DetectionOutputLayer
	Detection output layer. More...
class	Dict
	This class implements name-value dictionary, values are instances of DictValue. More...
struct	DictValue
	This struct stores the scalar value (or array) of one of the following type: double, cv::String or int64. More...
class	EinsumLayer
	This function performs array summation based on the Einstein summation convention. The function allows for concise expressions of various mathematical operations using subscripts. More...
class	EltwiseLayer
	Element wise operation on inputs. More...
class	EltwiseLayerInt8
class	ELULayer
class	ErfLayer
class	ExpandLayer
class	ExpLayer
class	FlattenLayer
class	FloorLayer
class	FlowWarpLayer
class	GatherElementsLayer
	GatherElements layer GatherElements takes two inputs data and indices of the same rank r >= 1 and an optional attribute axis and works such that: output[i][j][k] = data[index[i][j][k]][j][k] if axis = 0 and r = 3 output[i][j][k] = data[i][index[i][j][k]][k] if axis = 1 and r = 3 output[i][j][k] = data[i][j][index[i][j][k]] if axis = 2 and r = 3. More...
class	GatherLayer
	Gather layer. More...
class	GeluApproximationLayer
class	GeluLayer
class	GemmLayer
class	GroupNormLayer
class	GRULayer
	GRU recurrent one-layer. More...
class	HardSigmoidLayer
class	HardSwishLayer
struct	Image2BlobParams
	Processing params of image to blob. More...
class	InnerProductLayer
class	InnerProductLayerInt8
class	InstanceNormLayer
class	InterpLayer
	Bilinear resize layer from https://github.com/cdmh/deeplab-public-ver2. More...
class	KeypointsModel
	This class represents high-level API for keypoints models. More...
class	Layer
	This interface class allows to build new Layers - are building blocks of networks. More...
class	LayerFactory
	Layer factory allows to create instances of registered layers. More...
class	LayerNormLayer
class	LayerParams
	This class provides all data needed to initialize layer. More...
class	LogLayer
class	LRNLayer
class	LSTMLayer
	LSTM recurrent layer. More...
class	MatMulLayer
class	MaxUnpoolLayer
class	MishLayer
class	Model
	This class is presented high-level API for neural networks. More...
class	MVNLayer
class	NaryEltwiseLayer
class	Net
	This class allows to create and manipulate comprehensive artificial neural networks. More...
class	NormalizeBBoxLayer
	\( L_p \) - normalization layer. More...
class	NotLayer
class	PaddingLayer
	Adds extra values for specific axes. More...
class	PermuteLayer
class	PoolingLayer
class	PoolingLayerInt8
class	PowerLayer
class	PriorBoxLayer
class	ProposalLayer
class	QuantizeLayer
class	ReciprocalLayer
class	ReduceLayer
class	RegionLayer
class	ReLU6Layer
class	ReLULayer
class	ReorgLayer
class	RequantizeLayer
class	ReshapeLayer
class	ResizeLayer
	Resize input 4-dimensional blob by nearest neighbor or bilinear strategy. More...
class	RNNLayer
	Classical recurrent layer. More...
class	RoundLayer
class	ScaleLayer
class	ScaleLayerInt8
class	ScatterLayer
class	ScatterNDLayer
class	SegmentationModel
	This class represents high-level API for segmentation models. More...
class	SeluLayer
class	ShiftLayer
class	ShiftLayerInt8
class	ShrinkLayer
class	ShuffleChannelLayer
class	SigmoidLayer
class	SignLayer
class	SinhLayer
class	SinLayer
class	SliceLayer
class	SoftmaxLayer
class	SoftmaxLayerInt8
class	SoftplusLayer
class	SoftsignLayer
class	SplitLayer
class	SqrtLayer
class	SwishLayer
class	TanHLayer
class	TanLayer
class	TextDetectionModel
	Base class for text detection networks. More...
class	TextDetectionModel_DB
	This class represents high-level API for text detection DL networks compatible with DB model. More...
class	TextDetectionModel_EAST
	This class represents high-level API for text detection DL networks compatible with EAST model. More...
class	TextRecognitionModel
	This class represents high-level API for text recognition networks. More...
class	ThresholdedReluLayer
class	TileLayer

Typedefs
typedef std::map< std::string, std::vector< LayerFactory::Constructor > >	LayerFactory_Impl
typedef std::vector< int >	MatShape

Enumerations
enum	Backend {   DNN_BACKEND_DEFAULT = 0 ,   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 }
	Enum of computation backends supported by layers. More...
enum	DataLayout {   DNN_LAYOUT_UNKNOWN = 0 ,   DNN_LAYOUT_ND = 1 ,   DNN_LAYOUT_NCHW = 2 ,   DNN_LAYOUT_NCDHW = 3 ,   DNN_LAYOUT_NHWC = 4 ,   DNN_LAYOUT_NDHWC = 5 ,   DNN_LAYOUT_PLANAR = 6 }
	Enum of data layout for model inference. More...
enum	ImagePaddingMode {   DNN_PMODE_NULL = 0 ,   DNN_PMODE_CROP_CENTER = 1 ,   DNN_PMODE_LETTERBOX = 2 }
	Enum of image processing mode. To facilitate the specialization pre-processing requirements of the dnn model. For example, the letter box often used in the Yolo series of models. More...
enum class	SoftNMSMethod {   SoftNMSMethod::SOFTNMS_LINEAR = 1 ,   SoftNMSMethod::SOFTNMS_GAUSSIAN = 2 }
	Enum of Soft NMS methods. More...
enum	Target {   DNN_TARGET_CPU = 0 ,   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 }
	Enum of target devices for computations. More...

Functions
Mat	blobFromImage (InputArray image, double scalefactor=1.0, const Size &size=Size(), const Scalar &mean=Scalar(), bool swapRB=false, bool crop=false, int ddepth=CV_32F)
	Creates 4-dimensional blob from image. Optionally resizes and crops image from center, subtract mean values, scales values by scalefactor, swap Blue and Red channels.
void	blobFromImage (InputArray image, OutputArray blob, double scalefactor=1.0, const Size &size=Size(), const Scalar &mean=Scalar(), bool swapRB=false, bool crop=false, int ddepth=CV_32F)
	Creates 4-dimensional blob from image.
Mat	blobFromImages (InputArrayOfArrays images, double scalefactor=1.0, Size size=Size(), const Scalar &mean=Scalar(), bool swapRB=false, bool crop=false, int ddepth=CV_32F)
	Creates 4-dimensional blob from series of images. Optionally resizes and crops images from center, subtract mean values, scales values by scalefactor, swap Blue and Red channels.
void	blobFromImages (InputArrayOfArrays images, OutputArray blob, double scalefactor=1.0, Size size=Size(), const Scalar &mean=Scalar(), bool swapRB=false, bool crop=false, int ddepth=CV_32F)
	Creates 4-dimensional blob from series of images.
Mat	blobFromImagesWithParams (InputArrayOfArrays images, const Image2BlobParams &param=Image2BlobParams())
	Creates 4-dimensional blob from series of images with given params.
void	blobFromImagesWithParams (InputArrayOfArrays images, OutputArray blob, const Image2BlobParams &param=Image2BlobParams())
Mat	blobFromImageWithParams (InputArray image, const Image2BlobParams &param=Image2BlobParams())
	Creates 4-dimensional blob from image with given params.
void	blobFromImageWithParams (InputArray image, OutputArray blob, const Image2BlobParams &param=Image2BlobParams())
static MatShape	concat (const MatShape &a, const MatShape &b)
void	enableModelDiagnostics (bool isDiagnosticsMode)
	Enables detailed logging of the DNN model loading with CV DNN API.
std::vector< std::pair< Backend, Target > >	getAvailableBackends ()
std::vector< Target >	getAvailableTargets (dnn::Backend be)
cv::String	getInferenceEngineBackendType ()
	Returns Inference Engine internal backend API.
cv::String	getInferenceEngineCPUType ()
	Returns Inference Engine CPU type.
cv::String	getInferenceEngineVPUType ()
	Returns Inference Engine VPU type.
LayerFactory_Impl &	getLayerFactoryImpl ()
Mutex &	getLayerFactoryMutex ()
	Get the mutex guarding LayerFactory_Impl, see getLayerFactoryImpl() function.
static Mat	getPlane (const Mat &m, int n, int cn)
void	imagesFromBlob (const cv::Mat &blob_, OutputArrayOfArrays images_)
	Parse a 4D blob and output the images it contains as 2D arrays through a simpler data structure (std::vector<cv::Mat>).
static bool	isAllOnes (const MatShape &inputShape, int startPos, int endPos)
void	NMSBoxes (const std::vector< Rect > &bboxes, const std::vector< float > &scores, const float score_threshold, const float nms_threshold, std::vector< int > &indices, const float eta=1.f, const int top_k=0)
	Performs non maximum suppression given boxes and corresponding scores.
void	NMSBoxes (const std::vector< Rect2d > &bboxes, const std::vector< float > &scores, const float score_threshold, const float nms_threshold, std::vector< int > &indices, const float eta=1.f, const int top_k=0)
void	NMSBoxes (const std::vector< RotatedRect > &bboxes, const std::vector< float > &scores, const float score_threshold, const float nms_threshold, std::vector< int > &indices, const float eta=1.f, const int top_k=0)
void	NMSBoxesBatched (const std::vector< Rect > &bboxes, const std::vector< float > &scores, const std::vector< int > &class_ids, const float score_threshold, const float nms_threshold, std::vector< int > &indices, const float eta=1.f, const int top_k=0)
	Performs batched non maximum suppression on given boxes and corresponding scores across different classes.
void	NMSBoxesBatched (const std::vector< Rect2d > &bboxes, const std::vector< float > &scores, const std::vector< int > &class_ids, const float score_threshold, const float nms_threshold, std::vector< int > &indices, const float eta=1.f, const int top_k=0)
static int	normalize_axis (int axis, const MatShape &shape)
static int	normalize_axis (int axis, int dims)
	Converts axis from [-dims; dims) (similar to Python's slice notation) to [0; dims) range.
static Range	normalize_axis_range (const Range &r, int axisSize)
template<typename _Tp >
static std::ostream &	operator<< (std::ostream &out, const std::vector< _Tp > &shape)
template<typename _Tp >
static void	print (const std::vector< _Tp > &shape, const String &name="")
Net	readNet (const String &framework, const std::vector< uchar > &bufferModel, const std::vector< uchar > &bufferConfig=std::vector< uchar >())
	Read deep learning network represented in one of the supported formats.
Net	readNet (CV_WRAP_FILE_PATH const String &model, CV_WRAP_FILE_PATH const String &config="", const String &framework="")
	Read deep learning network represented in one of the supported formats.
Net	readNetFromCaffe (const char *bufferProto, size_t lenProto, const char *bufferModel=NULL, size_t lenModel=0)
	Reads a network model stored in Caffe model in memory.
Net	readNetFromCaffe (const std::vector< uchar > &bufferProto, const std::vector< uchar > &bufferModel=std::vector< uchar >())
	Reads a network model stored in Caffe model in memory.
Net	readNetFromCaffe (CV_WRAP_FILE_PATH const String &prototxt, CV_WRAP_FILE_PATH const String &caffeModel=String())
	Reads a network model stored in Caffe framework's format.
Net	readNetFromDarknet (const char *bufferCfg, size_t lenCfg, const char *bufferModel=NULL, size_t lenModel=0)
	Reads a network model stored in Darknet model files.
Net	readNetFromDarknet (const std::vector< uchar > &bufferCfg, const std::vector< uchar > &bufferModel=std::vector< uchar >())
	Reads a network model stored in Darknet model files.
Net	readNetFromDarknet (CV_WRAP_FILE_PATH const String &cfgFile, CV_WRAP_FILE_PATH const String &darknetModel=String())
	Reads a network model stored in Darknet model files.
Net	readNetFromModelOptimizer (const std::vector< uchar > &bufferModelConfig, const std::vector< uchar > &bufferWeights)
	Load a network from Intel's Model Optimizer intermediate representation.
Net	readNetFromModelOptimizer (const uchar *bufferModelConfigPtr, size_t bufferModelConfigSize, const uchar *bufferWeightsPtr, size_t bufferWeightsSize)
	Load a network from Intel's Model Optimizer intermediate representation.
Net	readNetFromModelOptimizer (CV_WRAP_FILE_PATH const String &xml, CV_WRAP_FILE_PATH const String &bin="")
	Load a network from Intel's Model Optimizer intermediate representation.
Net	readNetFromONNX (const char *buffer, size_t sizeBuffer)
	Reads a network model from ONNX in-memory buffer.
Net	readNetFromONNX (const std::vector< uchar > &buffer)
	Reads a network model from ONNX in-memory buffer.
Net	readNetFromONNX (CV_WRAP_FILE_PATH const String &onnxFile)
	Reads a network model ONNX.
Net	readNetFromTensorflow (const char *bufferModel, size_t lenModel, const char *bufferConfig=NULL, size_t lenConfig=0)
	Reads a network model stored in TensorFlow framework's format.
Net	readNetFromTensorflow (const std::vector< uchar > &bufferModel, const std::vector< uchar > &bufferConfig=std::vector< uchar >())
	Reads a network model stored in TensorFlow framework's format.
Net	readNetFromTensorflow (CV_WRAP_FILE_PATH const String &model, CV_WRAP_FILE_PATH const String &config=String())
	Reads a network model stored in TensorFlow framework's format.
Net	readNetFromTFLite (const char *bufferModel, size_t lenModel)
	Reads a network model stored in TFLite framework's format.
Net	readNetFromTFLite (const std::vector< uchar > &bufferModel)
	Reads a network model stored in TFLite framework's format.
Net	readNetFromTFLite (CV_WRAP_FILE_PATH const String &model)
	Reads a network model stored in TFLite framework's format.
Net	readNetFromTorch (CV_WRAP_FILE_PATH const String &model, bool isBinary=true, bool evaluate=true)
	Reads a network model stored in Torch7 framework's format.
Mat	readTensorFromONNX (CV_WRAP_FILE_PATH const String &path)
	Creates blob from .pb file.
Mat	readTorchBlob (const String &filename, bool isBinary=true)
	Loads blob which was serialized as torch.Tensor object of Torch7 framework.
void	releaseHDDLPlugin ()
	Release a HDDL plugin.
void	resetMyriadDevice ()
	Release a Myriad device (binded by OpenCV).
cv::String	setInferenceEngineBackendType (const cv::String &newBackendType)
	Specify Inference Engine internal backend API.
static MatShape	shape (const int *dims, const int n)
static MatShape	shape (const Mat &mat)
static MatShape	shape (const MatSize &sz)
static MatShape	shape (const UMat &mat)
static MatShape	shape (int a0, int a1=-1, int a2=-1, int a3=-1)
void	shrinkCaffeModel (CV_WRAP_FILE_PATH const String &src, CV_WRAP_FILE_PATH const String &dst, const std::vector< String > &layersTypes=std::vector< String >())
	Convert all weights of Caffe network to half precision floating point.
void	skipModelImport (bool skip)
	Skip model import after diagnostic run in readNet() functions.
static Mat	slice (const Mat &m, const _Range &r0)
static Mat	slice (const Mat &m, const _Range &r0, const _Range &r1)
static Mat	slice (const Mat &m, const _Range &r0, const _Range &r1, const _Range &r2)
static Mat	slice (const Mat &m, const _Range &r0, const _Range &r1, const _Range &r2, const _Range &r3)
void	softNMSBoxes (const std::vector< Rect > &bboxes, const std::vector< float > &scores, std::vector< float > &updated_scores, const float score_threshold, const float nms_threshold, std::vector< int > &indices, size_t top_k=0, const float sigma=0.5, SoftNMSMethod method=SoftNMSMethod::SOFTNMS_GAUSSIAN)
	Performs soft non maximum suppression given boxes and corresponding scores. Reference: https://arxiv.org/abs/1704.04503.
template<typename _Tp >
static std::string	toString (const std::vector< _Tp > &shape, const String &name="")
static int	total (const Mat &mat, int start=-1, int end=-1)
static int	total (const MatShape &shape, int start=-1, int end=-1)
void	writeTextGraph (CV_WRAP_FILE_PATH const String &model, CV_WRAP_FILE_PATH const String &output)
	Create a text representation for a binary network stored in protocol buffer format.

Function Documentation

concat()

static MatShape cv::dnn::concat ( const MatShape &  a, const MatShape &  b  )

inlinestatic

getInferenceEngineBackendType()

cv::String cv::dnn::getInferenceEngineBackendType

(

)

Returns Inference Engine internal backend API.

See values of CV_DNN_BACKEND_INFERENCE_ENGINE_* macros.

OPENCV_DNN_BACKEND_INFERENCE_ENGINE_TYPE runtime parameter (environment variable) is ignored since 4.6.0.

Deprecated:

getInferenceEngineCPUType()

cv::String cv::dnn::getInferenceEngineCPUType

(

)

Returns Inference Engine CPU type.

Specify OpenVINO plugin: CPU or ARM.

getInferenceEngineVPUType()

cv::String cv::dnn::getInferenceEngineVPUType

(

)

Returns Inference Engine VPU type.

See values of CV_DNN_INFERENCE_ENGINE_VPU_TYPE_* macros.

getPlane()

static Mat cv::dnn::getPlane ( const Mat &  m, int  n, int  cn  )

inlinestatic

Here is the call graph for this function:

isAllOnes()

static bool cv::dnn::isAllOnes ( const MatShape &  inputShape, int  startPos, int  endPos  )

inlinestatic

normalize_axis() [1/2]

static int cv::dnn::normalize_axis ( int  axis, const MatShape &  shape  )

inlinestatic

Here is the call graph for this function:

normalize_axis() [2/2]

static int cv::dnn::normalize_axis ( int  axis, int  dims  )

inlinestatic

Converts axis from [-dims; dims) (similar to Python's slice notation) to [0; dims) range.

normalize_axis_range()

static Range cv::dnn::normalize_axis_range ( const Range &  r, int  axisSize  )

inlinestatic

Here is the call graph for this function:

operator<<()

template<typename _Tp >

static std::ostream & cv::dnn::operator<< ( std::ostream &  out, const std::vector< _Tp > &  shape  )

inlinestatic

Here is the call graph for this function:

print()

template<typename _Tp >

static void cv::dnn::print ( const std::vector< _Tp > &  shape, const String &  name = ""  )

inlinestatic

Here is the call graph for this function:

releaseHDDLPlugin()

void cv::dnn::releaseHDDLPlugin

(

)

Release a HDDL plugin.

resetMyriadDevice()

void cv::dnn::resetMyriadDevice

(

)

Release a Myriad device (binded by OpenCV).

Single Myriad device cannot be shared across multiple processes which uses Inference Engine's Myriad plugin.

setInferenceEngineBackendType()

cv::String cv::dnn::setInferenceEngineBackendType

(

const cv::String &

newBackendType

)

Specify Inference Engine internal backend API.

See values of CV_DNN_BACKEND_INFERENCE_ENGINE_* macros.

Returns

previous value of internal backend API

Deprecated:

shape() [1/5]

static MatShape cv::dnn::shape ( const int *  dims, const int  n  )

inlinestatic

Here is the call graph for this function:

shape() [2/5]

static MatShape cv::dnn::shape ( const Mat &  mat )

inlinestatic

Here is the call graph for this function:

shape() [3/5]

static MatShape cv::dnn::shape ( const MatSize &  sz )

inlinestatic

Here is the call graph for this function:

shape() [4/5]

static MatShape cv::dnn::shape ( const UMat &  mat )

inlinestatic

Here is the call graph for this function:

shape() [5/5]

static MatShape cv::dnn::shape ( int  a0, int  a1 = -1, int  a2 = -1, int  a3 = -1  )

inlinestatic

Here is the call graph for this function:

skipModelImport()

void cv::dnn::skipModelImport

(

bool

skip

)

Skip model import after diagnostic run in readNet() functions.

Parameters

[in]

skip

Indicates whether to skip the import.

This is an internal OpenCV function not intended for users.

slice() [1/4]

static Mat cv::dnn::slice ( const Mat &  m, const _Range &  r0  )

inlinestatic

Here is the call graph for this function:

slice() [2/4]

static Mat cv::dnn::slice ( const Mat &  m, const _Range &  r0, const _Range &  r1  )

inlinestatic

Here is the call graph for this function:

slice() [3/4]

static Mat cv::dnn::slice ( const Mat &  m, const _Range &  r0, const _Range &  r1, const _Range &  r2  )

inlinestatic

Here is the call graph for this function:

slice() [4/4]

static Mat cv::dnn::slice ( const Mat &  m, const _Range &  r0, const _Range &  r1, const _Range &  r2, const _Range &  r3  )

inlinestatic

Here is the call graph for this function:

toString()

template<typename _Tp >

static std::string cv::dnn::toString ( const std::vector< _Tp > &  shape, const String &  name = ""  )

inlinestatic

Here is the call graph for this function:

total() [1/2]

static int cv::dnn::total ( const Mat &  mat, int  start = -1, int  end = -1  )

inlinestatic

Here is the call graph for this function:

total() [2/2]

static int cv::dnn::total ( const MatShape &  shape, int  start = -1, int  end = -1  )

inlinestatic

Here is the call graph for this function: