Deep neural networks

All modules / dnn

Filter dnn

dnn_ActivationTypeclass

Activation type enumeration for dispatched activation function retrieval.

dnn_ArgKindclass

Indicate that an actual data is on CPU.

dnn_AutoPaddingclass

GatherND layer

dnn_Backendclass

Enum of computation backends supported by layers.

dnn_blobFromImagefunction

Creates 4-dimensional blob from image. Optionally resizes and crops `image` from center,

dnn_blobFromImagesfunction

Creates 4-dimensional blob from series of images. Optionally resizes and

dnn_blobFromImagesWithParamsfunction

Creates 4-dimensional blob from series of images with given params.

dnn_blobFromImagesWithParams1function

Creates 4-dimensional blob from series of images with given params.

dnn_blobFromImageWithParamsfunction

Creates 4-dimensional blob from image with given params.

dnn_blobFromImageWithParams1function

Creates 4-dimensional blob from image with given params.

dnn_ClassificationModelclass

This class represents high-level API for classification models.

dnn_DetectionModelclass

This class represents high-level API for object detection networks.

dnn_DictValueclass

This struct stores the scalar value (or array) of one of the following type: double, cv::String or int64.

dnn_EngineTypeclass

Native enum entries for dnn engine type. Pass an entry to enum-typed arguments or its .value to numeric arguments.

dnn_getAvailableTargetsfunction

Return an owned vector of execution targets available for a DNN backend in this build. CPU OpenCV inference is enabled.

dnn_getInferenceEngineBackendTypefunction

Returns Inference Engine internal backend API.

dnn_getInferenceEngineCPUTypefunction

Returns Inference Engine CPU type.

dnn_getInferenceEngineVPUTypefunction

Returns Inference Engine VPU type.

dnn_Image2BlobParamsclass

Processing params of image to blob.

dnn_ImagePaddingModeclass

Enum of image processing mode.

dnn_imagesFromBlobfunction

Parse a 4D blob and output the images it contains as 2D arrays through a simpler data structure

dnn_KeypointsModelclass

This class represents high-level API for keypoints models

dnn_Layerclass

This interface class allows to build new Layers - are building blocks of networks.

dnn_LayerParamsclass

A DNN layer's instance name, type, learned blobs and named configuration values.

dnn_LossReductionclass

Shared reduction enum for DNN loss layers

dnn_Modelclass

This class is presented high-level API for neural networks.

dnn_ModelFormatclass

Native enum entries for dnn model format. Pass an entry to enum-typed arguments or its .value to numeric arguments.

dnn_NaryEltwiseLayer_OPERATIONclass

Element wise operation on inputs

dnn_Netclass

This class allows to create and manipulate comprehensive artificial neural networks.

dnn_Net_readFromModelOptimizerfunction

Destructor frees the net only if there aren't references to the net anymore.

dnn_NMSBoxesfunction

Performs non maximum suppression given boxes and corresponding scores.

dnn_NMSBoxesBatchedfunction

Performs batched non maximum suppression on given boxes and corresponding scores across different classes.

dnn_NMSBoxesRotatedfunction

Performs non maximum suppression given boxes and corresponding scores.

dnn_ProfilingModeclass

Native enum entries for dnn profiling mode. Pass an entry to enum-typed arguments or its .value to numeric arguments.

dnn_readNetfunction

Read deep learning network represented in one of the supported formats.

dnn_readNet1function

Read deep learning network represented in one of the supported formats.

dnn_readNetFromModelOptimizerfunction

Load a network from Intel's Model Optimizer intermediate representation.

dnn_readNetFromModelOptimizer1function

Load a network from Intel's Model Optimizer intermediate representation.

dnn_readNetFromONNXfunction

Reads a network model <a href="https://onnx.ai/">ONNX</a>.

dnn_readNetFromONNX1function

Reads a network model from <a href="https://onnx.ai/">ONNX</a>

dnn_readNetFromTensorflowfunction

Reads a network model stored in <a href="https://www.tensorflow.org/">TensorFlow</a> framework's format.

dnn_readNetFromTensorflow1function

Reads a network model stored in <a href="https://www.tensorflow.org/">TensorFlow</a> framework's format.

dnn_readNetFromTFLitefunction

Reads a network model stored in <a href="https://www.tensorflow.org/lite">TFLite</a> framework's format.

dnn_readNetFromTFLite1function

Reads a network model stored in <a href="https://www.tensorflow.org/lite">TFLite</a> framework's format.

dnn_readTensorFromONNXfunction

Creates blob from .pb file.

dnn_Reduce2Layer_ReduceTypeclass

GatherElements layer

dnn_releaseHDDLPluginfunction

Release a HDDL plugin.

dnn_resetMyriadDevicefunction

Release a Myriad device (binded by OpenCV).

dnn_SegmentationModelclass

This class represents high-level API for segmentation models

dnn_setInferenceEngineBackendTypefunction

Specify Inference Engine internal backend API.

dnn_softNMSBoxesfunction

Performs soft non maximum suppression given boxes and corresponding scores.

dnn_SoftNMSMethodclass

Enum of Soft NMS methods.

dnn_Targetclass

Enum of target devices for computations.

dnn_TextDetectionModelclass

Base class for text detection networks

dnn_TextDetectionModel_DBclass

This class represents high-level API for text detection DL networks compatible with DB model.

dnn_TextDetectionModel_EASTclass

This class represents high-level API for text detection DL networks compatible with EAST model.

dnn_TextRecognitionModelclass

This class represents high-level API for text recognition networks.

dnn_Tokenizerclass

High-level tokenizer wrapper for DNN usage.

dnn_Tokenizer_loadfunction

Load a tokenizer from a model directory.

dnn_TracingModeclass

Native enum entries for dnn tracing mode. Pass an entry to enum-typed arguments or its .value to numeric arguments.

dnn_writeTextGraphfunction

Create a text representation for a binary network stored in protocol buffer format.

Constants

Use the named constant from cv. In particular, OpenCV 5 matrix type codes differ from OpenCV 4.

dnn_AUTO_PAD_NONE

dnn auto pad none constant (0), defined by OpenCV for cv::dnn.

dnn_AUTO_PAD_SAME_UPPER

dnn auto pad same upper constant (1), defined by OpenCV for cv::dnn.

dnn_AUTO_PAD_SAME_LOWER

dnn auto pad same lower constant (2), defined by OpenCV for cv::dnn.

dnn_AUTO_PAD_VALID

dnn auto pad valid constant (3), defined by OpenCV for cv::dnn.

dnn_Reduce2Layer_ReduceType_MAX

dnn reduce2 layer reduce type max constant (0), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_REDUCE2LAYER_REDUCE_TYPE_MAX

dnn reduce2 layer reduce type max constant (0), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_Reduce2Layer_ReduceType_MIN

dnn reduce2 layer reduce type min constant (1), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_REDUCE2LAYER_REDUCE_TYPE_MIN

dnn reduce2 layer reduce type min constant (1), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_Reduce2Layer_ReduceType_MEAN

dnn reduce2 layer reduce type mean constant (2), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_REDUCE2LAYER_REDUCE_TYPE_MEAN

dnn reduce2 layer reduce type mean constant (2), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_Reduce2Layer_ReduceType_SUM

dnn reduce2 layer reduce type sum constant (3), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_REDUCE2LAYER_REDUCE_TYPE_SUM

dnn reduce2 layer reduce type sum constant (3), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_Reduce2Layer_ReduceType_L1

dnn reduce2 layer reduce type l1 constant (4), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_REDUCE2LAYER_REDUCE_TYPE_L1

dnn reduce2 layer reduce type l1 constant (4), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_Reduce2Layer_ReduceType_L2

dnn reduce2 layer reduce type l2 constant (5), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_REDUCE2LAYER_REDUCE_TYPE_L2

dnn reduce2 layer reduce type l2 constant (5), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_Reduce2Layer_ReduceType_PROD

dnn reduce2 layer reduce type prod constant (6), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_REDUCE2LAYER_REDUCE_TYPE_PROD

dnn reduce2 layer reduce type prod constant (6), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_Reduce2Layer_ReduceType_SUM_SQUARE

dnn reduce2 layer reduce type sum square constant (7), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_REDUCE2LAYER_REDUCE_TYPE_SUM_SQUARE

dnn reduce2 layer reduce type sum square constant (7), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_Reduce2Layer_ReduceType_LOG_SUM

dnn reduce2 layer reduce type log sum constant (8), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_REDUCE2LAYER_REDUCE_TYPE_LOG_SUM

dnn reduce2 layer reduce type log sum constant (8), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_Reduce2Layer_ReduceType_LOG_SUM_EXP

dnn reduce2 layer reduce type log sum exp constant (9), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_REDUCE2LAYER_REDUCE_TYPE_LOG_SUM_EXP

dnn reduce2 layer reduce type log sum exp constant (9), defined by OpenCV for cv::dnn::Reduce2Layer::ReduceType.

dnn_ACTIV_NONE

dnn activ none constant (0), defined by OpenCV for cv::dnn.

dnn_ACTIV_MISH

dnn activ mish constant (0+1), defined by OpenCV for cv::dnn.

dnn_ACTIV_SWISH

dnn activ swish constant (0+2), defined by OpenCV for cv::dnn.

dnn_ACTIV_SIGMOID

dnn activ sigmoid constant (0+3), defined by OpenCV for cv::dnn.

dnn_ACTIV_TANH

dnn activ tanh constant (0+4), defined by OpenCV for cv::dnn.

dnn_ACTIV_ELU

dnn activ elu constant (0+5), defined by OpenCV for cv::dnn.

dnn_ACTIV_HARDSWISH

dnn activ hardswish constant (0+6), defined by OpenCV for cv::dnn.

dnn_ACTIV_HARDSIGMOID

dnn activ hardsigmoid constant (0+7), defined by OpenCV for cv::dnn.

dnn_ACTIV_GELU

dnn activ gelu constant (0+8), defined by OpenCV for cv::dnn.

dnn_ACTIV_GELU_APPROX

dnn activ gelu approx constant (0+9), defined by OpenCV for cv::dnn.

dnn_ACTIV_RELU

dnn activ relu constant (0+10), defined by OpenCV for cv::dnn.

dnn_ACTIV_CLIP

dnn activ clip constant (0+11), defined by OpenCV for cv::dnn.

dnn_NaryEltwiseLayer_OPERATION_AND

dnn nary eltwise layer operation and constant (0), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_AND

dnn nary eltwise layer operation and constant (0), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_EQUAL

dnn nary eltwise layer operation equal constant (0+1), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_EQUAL

dnn nary eltwise layer operation equal constant (0+1), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_GREATER

dnn nary eltwise layer operation greater constant (0+2), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_GREATER

dnn nary eltwise layer operation greater constant (0+2), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_GREATER_EQUAL

dnn nary eltwise layer operation greater equal constant (0+3), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_GREATER_EQUAL

dnn nary eltwise layer operation greater equal constant (0+3), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_LESS

dnn nary eltwise layer operation less constant (0+4), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_LESS

dnn nary eltwise layer operation less constant (0+4), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_LESS_EQUAL

dnn nary eltwise layer operation less equal constant (0+5), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_LESS_EQUAL

dnn nary eltwise layer operation less equal constant (0+5), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_OR

dnn nary eltwise layer operation or constant (0+6), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_OR

dnn nary eltwise layer operation or constant (0+6), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_POW

dnn nary eltwise layer operation pow constant (0+7), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_POW

dnn nary eltwise layer operation pow constant (0+7), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_XOR

dnn nary eltwise layer operation xor constant (0+8), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_XOR

dnn nary eltwise layer operation xor constant (0+8), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_BITSHIFT

dnn nary eltwise layer operation bitshift constant (0+9), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_BITSHIFT

dnn nary eltwise layer operation bitshift constant (0+9), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_MAX

dnn nary eltwise layer operation max constant (0+10), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_MAX

dnn nary eltwise layer operation max constant (0+10), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_MEAN

dnn nary eltwise layer operation mean constant (0+11), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_MEAN

dnn nary eltwise layer operation mean constant (0+11), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_MIN

dnn nary eltwise layer operation min constant (0+12), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_MIN

dnn nary eltwise layer operation min constant (0+12), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_MOD

Integer Mod. Reminder's sign = Divisor's sign.

dnn_NARY_ELTWISE_LAYER_OPERATION_MOD

Integer Mod. Reminder's sign = Divisor's sign.

dnn_NaryEltwiseLayer_OPERATION_FMOD

Floating-point Mod. Reminder's sign = Dividend's sign.

dnn_NARY_ELTWISE_LAYER_OPERATION_FMOD

Floating-point Mod. Reminder's sign = Dividend's sign.

dnn_NaryEltwiseLayer_OPERATION_PROD

Floating-point Mod. Reminder's sign = Dividend's sign.

dnn_NARY_ELTWISE_LAYER_OPERATION_PROD

Floating-point Mod. Reminder's sign = Dividend's sign.

dnn_NaryEltwiseLayer_OPERATION_SUB

dnn nary eltwise layer operation sub constant (0+16), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_SUB

dnn nary eltwise layer operation sub constant (0+16), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_SUM

dnn nary eltwise layer operation sum constant (0+17), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_SUM

dnn nary eltwise layer operation sum constant (0+17), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_ADD

dnn nary eltwise layer operation add constant (0+18), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_ADD

dnn nary eltwise layer operation add constant (0+18), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_DIV

dnn nary eltwise layer operation div constant (0+19), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_DIV

dnn nary eltwise layer operation div constant (0+19), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_WHERE

dnn nary eltwise layer operation where constant (0+20), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_WHERE

dnn nary eltwise layer operation where constant (0+20), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_BITWISE_AND

dnn nary eltwise layer operation bitwise and constant (0+21), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_BITWISE_AND

dnn nary eltwise layer operation bitwise and constant (0+21), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_BITWISE_OR

dnn nary eltwise layer operation bitwise or constant (0+22), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_BITWISE_OR

dnn nary eltwise layer operation bitwise or constant (0+22), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NaryEltwiseLayer_OPERATION_BITWISE_XOR

dnn nary eltwise layer operation bitwise xor constant (0+23), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_NARY_ELTWISE_LAYER_OPERATION_BITWISE_XOR

dnn nary eltwise layer operation bitwise xor constant (0+23), defined by OpenCV for cv::dnn::NaryEltwiseLayer::OPERATION.

dnn_LOSS_REDUCTION_NONE

dnn loss reduction none constant (0), defined by OpenCV for cv::dnn.

dnn_LOSS_REDUCTION_MEAN

dnn loss reduction mean constant (1), defined by OpenCV for cv::dnn.

dnn_LOSS_REDUCTION_SUM

dnn loss reduction sum constant (2), defined by OpenCV for cv::dnn.

dnn_DNN_BACKEND_DEFAULT

DNN_BACKEND_DEFAULT equals to OPENCV_DNN_BACKEND_DEFAULT, which can be defined using CMake or a configuration parameter

dnn_DNN_BACKEND_INFERENCE_ENGINE

internal - use DNN_BACKEND_INFERENCE_ENGINE + setInferenceEngineBackendType()

dnn_DNN_BACKEND_OPENCV

Intel OpenVINO computational backend, supported targets: CPU, OPENCL, OPENCL_FP16, MYRIAD, HDDL, NPU

Note: Tutorial how to build OpenCV with OpenVINO: tutorial_dnn_openvino

dnn_DNN_BACKEND_VKCOM

dnn dnn backend vkcom constant (2+2), defined by OpenCV for cv::dnn.

dnn_DNN_BACKEND_CUDA

dnn dnn backend cuda constant (2+3), defined by OpenCV for cv::dnn.

dnn_DNN_BACKEND_WEBNN

dnn dnn backend webnn constant (2+4), defined by OpenCV for cv::dnn.

dnn_DNN_BACKEND_TIMVX

dnn dnn backend timvx constant (2+5), defined by OpenCV for cv::dnn.

dnn_DNN_BACKEND_CANN

dnn dnn backend cann constant (2+6), defined by OpenCV for cv::dnn.

dnn_DNN_TARGET_CPU

dnn dnn target cpu constant (0), defined by OpenCV for cv::dnn.

dnn_DNN_TARGET_OPENCL

dnn dnn target opencl constant (0+1), defined by OpenCV for cv::dnn.

dnn_DNN_TARGET_OPENCL_FP16

dnn dnn target opencl fp16 constant (0+2), defined by OpenCV for cv::dnn.

dnn_DNN_TARGET_MYRIAD

dnn dnn target myriad constant (0+3), defined by OpenCV for cv::dnn.

dnn_DNN_TARGET_VULKAN

dnn dnn target vulkan constant (0+4), defined by OpenCV for cv::dnn.

dnn_DNN_TARGET_FPGA

FPGA device with CPU fallbacks using Inference Engine's Heterogeneous plugin.

dnn_DNN_TARGET_CUDA

FPGA device with CPU fallbacks using Inference Engine's Heterogeneous plugin.

dnn_DNN_TARGET_CUDA_FP16

dnn dnn target cuda fp16 constant (0+7), defined by OpenCV for cv::dnn.

dnn_DNN_TARGET_HDDL

dnn dnn target hddl constant (0+8), defined by OpenCV for cv::dnn.

dnn_DNN_TARGET_NPU

dnn dnn target npu constant (0+9), defined by OpenCV for cv::dnn.

dnn_DNN_TARGET_CPU_FP16

Only the ARM platform is supported. Low precision computing, accelerate model inference.

dnn_DNN_TRACE_NONE

Don't trace anything

dnn_DNN_TRACE_ALL

Print all executed operations along with the output tensors, more or less compatible with ONNX Runtime

dnn_DNN_TRACE_OP

Print all executed operations. Types and shapes of all inputs and outputs are printed, but the content is not.

dnn_DNN_PROFILE_NONE

Don't do any profiling

dnn_DNN_PROFILE_SUMMARY

Collect the summary statistics by layer type (e.g. all "Conv2D" or all "Add") and print it in the end, sorted by the execution time (most expensive layers first). Note that it may introduce some overhead and cause slowdown, especially in the case of non-CPU backends.

dnn_DNN_PROFILE_DETAILED

Print execution time of each single layer. Note that it may introduce some overhead and cause slowdown, especially in the case of non-CPU backends.

dnn_DNN_MODEL_GENERIC

Some generic model format

dnn_DNN_MODEL_ONNX

dnn_DNN_MODEL_ONNX: ONNX model

dnn_DNN_MODEL_TF

dnn_DNN_MODEL_TF: TF model

dnn_DNN_MODEL_TFLITE

dnn_DNN_MODEL_TFLITE: TFLite model

dnn_DNN_ARG_EMPTY

valid only for Arg.idx==0. It's "no-arg"

dnn_DNN_ARG_CONST

a constant argument.

dnn_DNN_ARG_INPUT

input of the whole model. Before Net::forward() or in Net::forward() all inputs must be set

dnn_DNN_ARG_OUTPUT

output of the model.

dnn_DNN_ARG_TEMP

intermediate result, a result of some operation and input to some other operation(s).

dnn_DNN_ARG_PATTERN

not used for now

dnn_ENGINE_CLASSIC

Force use the old dnn engine similar to 4.x branch

dnn_ENGINE_NEW

Force use the new dnn engine. The engine does not support non CPU back-ends for now.

dnn_ENGINE_AUTO

Try to use the new engine and then fall back to the classic version.

dnn_ENGINE_ORT

Try to use ONNX Runtime wrapper (ONNX only, requires build with WITH_ONNXRUNTIME=ON).

dnn_DNN_PMODE_NULL

!< Default. Resize to required input size without extra processing.

dnn_DNN_PMODE_CROP_CENTER

!< Image will be cropped after resize.

dnn_DNN_PMODE_LETTERBOX

!< Resize image to the desired size while preserving the aspect ratio of original image.

dnn_SoftNMSMethod_SOFTNMS_LINEAR

dnn soft nmsmethod softnms linear constant (1), defined by OpenCV for cv::dnn::SoftNMSMethod.

dnn_SOFT_NMSMETHOD_SOFTNMS_LINEAR

dnn soft nmsmethod softnms linear constant (1), defined by OpenCV for cv::dnn::SoftNMSMethod.

dnn_SoftNMSMethod_SOFTNMS_GAUSSIAN

dnn soft nmsmethod softnms gaussian constant (2), defined by OpenCV for cv::dnn::SoftNMSMethod.

dnn_SOFT_NMSMETHOD_SOFTNMS_GAUSSIAN

dnn soft nmsmethod softnms gaussian constant (2), defined by OpenCV for cv::dnn::SoftNMSMethod.