This structure provides functions that fill inference parameters for "OpenVINO Toolkit" model. More...

#include <opencv2/gapi/infer/ov.hpp>

Collaboration diagram for cv::gapi::ov::Params< Net >:

Public Member Functions
	Params (const std::string &blob_path, const std::string &device)

	Params (const std::string &model_path, const std::string &bin_path, const std::string &device)
	Class constructor.

GBackend	backend () const

Params< Net > &	cfgClampOutputs (bool flag=true)
	Enables or disables clamping of model outputs in the PrePostProcessor.

Params< Net > &	cfgEnsureNamedTensors (bool flag=true)
	Ensures the model has named tensors.

Params< Net > &	cfgInputLayers (const std::vector< std::string > &layer_names)
	Specifies sequence of network input layers names for inference.

Params< Net > &	cfgInputModelLayout (detail::AttrMap< std::string > layout_map)

Params< Net > &	cfgInputModelLayout (std::string layout)
	Specifies model layout for an input layer.

Params< Net > &	cfgInputTensorLayout (detail::AttrMap< std::string > layout_map)

Params< Net > &	cfgInputTensorLayout (std::string layout)
	Specifies tensor layout for an input layer.

Params< Net > &	cfgMean (detail::AttrMap< std::vector< float > > mean_map)

Params< Net > &	cfgMean (std::vector< float > mean_values)
	Specifies mean values for preprocessing.

Params< Net > &	cfgNumRequests (const size_t nireq)
	Specifies number of asynchronous inference requests.

Params< Net > &	cfgOutputLayers (const std::vector< std::string > &layer_names)
	Specifies sequence of network output layers names for inference.

Params< Net > &	cfgOutputModelLayout (detail::AttrMap< std::string > layout_map)

Params< Net > &	cfgOutputModelLayout (std::string layout)
	Specifies model layout for an output layer.

Params< Net > &	cfgOutputTensorLayout (detail::AttrMap< std::string > layout_map)

Params< Net > &	cfgOutputTensorLayout (std::string layout)
	Specifies tensor layout for an output layer.

Params< Net > &	cfgOutputTensorPrecision (detail::AttrMap< int > precision_map)

Params< Net > &	cfgOutputTensorPrecision (int precision)
	Specifies tensor precision for an output layer.

Params< Net > &	cfgPluginConfig (const detail::ParamDesc::PluginConfigT &config)
	Specifies OpenVINO plugin configuration.

Params< Net > &	cfgReshape (detail::AttrMap< std::vector< size_t > > new_shape_map)

Params< Net > &	cfgReshape (std::vector< size_t > new_shape)
	Specifies the new shape for input layers.

Params< Net > &	cfgResize (detail::AttrMap< int > interpolation)

Params< Net > &	cfgResize (int interpolation)
	Specifies resize interpolation algorithm.

Params< Net > &	cfgScale (detail::AttrMap< std::vector< float > > scale_map)

Params< Net > &	cfgScale (std::vector< float > scale_values)
	Specifies scale values for preprocessing.

cv::util::any	params () const

std::string	tag () const

Protected Attributes
detail::ParamDesc	m_desc

Detailed Description

template<typename Net>
struct cv::gapi::ov::Params< Net >

This structure provides functions that fill inference parameters for "OpenVINO Toolkit" model.

Constructor & Destructor Documentation

◆ Params() [1/2]

template<typename Net >

cv::gapi::ov::Params< Net >::Params	(	const std::string &	model_path,
		const std::string &	bin_path,
		const std::string &	device )

inline

Class constructor.

Constructs Params based on model information and specifies default values for other inference description parameters. Model is loaded and compiled using "OpenVINO Toolkit".

Parameters

model_path	Path to a model.
bin_path	Path to a data file. For IR format (.bin): If path is empty, will try to read a bin file with the same name as xml. If the bin file with the same name is not found, will load IR without weights. For PDPD (.pdmodel) and ONNX (*.onnx) formats bin_path isn't used.
device	target device to use.

◆ Params() [2/2]

template<typename Net >

cv::gapi::ov::Params< Net >::Params	(	const std::string &	blob_path,
		const std::string &	device )

inline

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts. Use this constructor to work with pre-compiled network. Model is imported from a pre-compiled blob.

Parameters

blob_path	path to the compiled model (*.blob).
device	target device to use.

Member Function Documentation

◆ backend()

template<typename Net >

GBackend cv::gapi::ov::Params< Net >::backend ( ) const

inline

Here is the call graph for this function:

◆ cfgClampOutputs()

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgClampOutputs ( bool flag = true )

inline

Enables or disables clamping of model outputs in the PrePostProcessor.

By default, output values are clamped to the valid range for the output precision by the device or plugin. Enabling this option moves clamping to the PrePostProcessor stage.

Note: This feature is only available with OpenVINO 2025.2 and newer.

Parameters

flag	If true, clamping is performed in the PrePostProcessor; otherwise, it is handled by the device or plugin.

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgEnsureNamedTensors()

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgEnsureNamedTensors ( bool flag = true )

inline

Ensures the model has named tensors.

This function is used to ensure that all tensors in the model have names. It goes through all input and output nodes of the model and sets the names if they are not set. This is neccessary for models with nameless tensors.

If a tensor does not have a name, it will be assigned a default name based on the producer node's friendly name. If the producer node has multiple outputs, the name will be in the form "node_name:N", where N is the output index.

Parameters

flag	If true, then it guarantees that all tensors will have names.

Returns: reference to this parameter structure.

◆ cfgInputLayers()

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgInputLayers ( const std::vector< std::string > & layer_names )

inline

Specifies sequence of network input layers names for inference.

The function is used to associate cv::gapi::infer<> inputs with the model inputs. Number of names has to match the number of network inputs as defined in G_API_NET(). In case a network has only single input layer, there is no need to specify name manually.

Parameters

layer_names std::array<std::string, N> where N is the number of inputs as defined in the G_API_NET. Contains names of input layers.

Returns: reference to this parameter structure.

◆ cfgInputModelLayout() [1/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgInputModelLayout ( detail::AttrMap< std::string > layout_map )

inline

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Parameters

layout_map Map of pairs: name of corresponding input layer and its model layout ("NCHW", "NHWC", etc)

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgInputModelLayout() [2/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgInputModelLayout ( std::string layout )

inline

Specifies model layout for an input layer.

The function is used to set model layout for an input layer.

Parameters

layout Model layout ("NCHW", "NHWC", etc) will be applied to all input layers.

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgInputTensorLayout() [1/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgInputTensorLayout ( detail::AttrMap< std::string > layout_map )

inline

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Parameters

layout_map Map of pairs: name of corresponding input layer and its tensor layout represented in std::string ("NCHW", "NHWC", etc)

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgInputTensorLayout() [2/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgInputTensorLayout ( std::string layout )

inline

Specifies tensor layout for an input layer.

The function is used to set tensor layout for an input layer.

Parameters

layout Tensor layout ("NCHW", "NWHC", etc) will be applied to all input layers.

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgMean() [1/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgMean ( detail::AttrMap< std::vector< float > > mean_map )

inline

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Parameters

mean_map Map of pairs: name of corresponding input layer and its mean values.

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgMean() [2/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgMean ( std::vector< float > mean_values )

inline

Specifies mean values for preprocessing.

The function is used to set mean values for input layer preprocessing.

Parameters

mean_values Float vector contains mean values

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgNumRequests()

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgNumRequests ( const size_t nireq )

inline

Specifies number of asynchronous inference requests.

Parameters

nireq Number of inference asynchronous requests.

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgOutputLayers()

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgOutputLayers ( const std::vector< std::string > & layer_names )

inline

Specifies sequence of network output layers names for inference.

The function is used to associate cv::gapi::infer<> outputs with the model outputs. Number of names has to match the number of network outputs as defined in G_API_NET(). In case a network has only single output layer, there is no need to specify name manually.

Parameters

layer_names std::array<std::string, N> where N is the number of outputs as defined in the G_API_NET. Contains names of output layers.

Returns: reference to this parameter structure.

◆ cfgOutputModelLayout() [1/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgOutputModelLayout ( detail::AttrMap< std::string > layout_map )

inline

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Parameters

layout_map Map of pairs: name of corresponding output layer and its model layout ("NCHW", "NHWC", etc)

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgOutputModelLayout() [2/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgOutputModelLayout ( std::string layout )

inline

Specifies model layout for an output layer.

The function is used to set model layout for an output layer.

Parameters

layout Model layout ("NCHW", "NHWC", etc) will be applied to all output layers.

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgOutputTensorLayout() [1/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgOutputTensorLayout ( detail::AttrMap< std::string > layout_map )

inline

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Parameters

layout_map Map of pairs: name of corresponding output layer and its tensor layout represented in std::string ("NCHW", "NHWC", etc)

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgOutputTensorLayout() [2/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgOutputTensorLayout ( std::string layout )

inline

Specifies tensor layout for an output layer.

The function is used to set tensor layout for an output layer.

Parameters

layout Tensor layout ("NCHW", "NWHC", etc) will be applied to all output layers.

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgOutputTensorPrecision() [1/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgOutputTensorPrecision ( detail::AttrMap< int > precision_map )

inline

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Parameters

precision_map Map of pairs: name of corresponding output layer and its precision in OpenCV format (CV_8U, CV_32F, ...)

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgOutputTensorPrecision() [2/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgOutputTensorPrecision ( int precision )

inline

Specifies tensor precision for an output layer.

The function is used to set tensor precision for an output layer..

Parameters

precision Precision in OpenCV format (CV_8U, CV_32F, ...) will be applied to all output layers.

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgPluginConfig()

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgPluginConfig ( const detail::ParamDesc::PluginConfigT & config )

inline

Specifies OpenVINO plugin configuration.

The function is used to set configuration for OpenVINO plugin. Some parameters can be different for each plugin. Please follow https://docs.openvinotoolkit.org/latest/index.html to check information about specific plugin.

Parameters

config Map of pairs: (config parameter name, config parameter value).

Returns: reference to this parameter structure.

◆ cfgReshape() [1/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgReshape ( detail::AttrMap< std::vector< size_t > > new_shape_map )

inline

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Parameters

new_shape_map Map of pairs: name of corresponding output layer and its new shape.

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgReshape() [2/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgReshape ( std::vector< size_t > new_shape )

inline

Specifies the new shape for input layers.

The function is used to set new shape for input layers.

Parameters

new_shape New shape will be applied to all input layers.

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgResize() [1/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgResize ( detail::AttrMap< int > interpolation )

inline

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Parameters

interpolation Map of pairs: name of corresponding input layer and its resize algorithm.

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgResize() [2/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgResize ( int interpolation )

inline

Specifies resize interpolation algorithm.

The function is used to configure resize preprocessing for input layer.

Parameters

interpolation Resize interpolation algorithm. Supported algorithms: INTER_NEAREST, INTER_LINEAR, INTER_CUBIC.

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgScale() [1/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgScale ( detail::AttrMap< std::vector< float > > scale_map )

inline

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Parameters

scale_map Map of pairs: name of corresponding input layer and its mean values.

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ cfgScale() [2/2]

template<typename Net >

Params< Net > & cv::gapi::ov::Params< Net >::cfgScale ( std::vector< float > scale_values )

inline

Specifies scale values for preprocessing.

The function is used to set scale values for input layer preprocessing.

Parameters

scale_values Float vector contains scale values

Returns: reference to this parameter structure.

Here is the call graph for this function:

◆ params()

template<typename Net >

cv::util::any cv::gapi::ov::Params< Net >::params ( ) const

inline

◆ tag()

template<typename Net >

std::string cv::gapi::ov::Params< Net >::tag ( ) const

inline

Member Data Documentation

◆ m_desc

template<typename Net >

detail::ParamDesc cv::gapi::ov::Params< Net >::m_desc

protected

The documentation for this struct was generated from the following file:

opencv2/gapi/infer/ov.hpp

Public Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

◆ Params() [1/2]

◆ Params() [2/2]

Member Function Documentation

◆ backend()

◆ cfgClampOutputs()

◆ cfgEnsureNamedTensors()

◆ cfgInputLayers()

◆ cfgInputModelLayout() [1/2]

◆ cfgInputModelLayout() [2/2]

◆ cfgInputTensorLayout() [1/2]

◆ cfgInputTensorLayout() [2/2]

◆ cfgMean() [1/2]

◆ cfgMean() [2/2]

◆ cfgNumRequests()

◆ cfgOutputLayers()

◆ cfgOutputModelLayout() [1/2]

◆ cfgOutputModelLayout() [2/2]

◆ cfgOutputTensorLayout() [1/2]

◆ cfgOutputTensorLayout() [2/2]

◆ cfgOutputTensorPrecision() [1/2]

◆ cfgOutputTensorPrecision() [2/2]

◆ cfgPluginConfig()

◆ cfgReshape() [1/2]

◆ cfgReshape() [2/2]

◆ cfgResize() [1/2]

◆ cfgResize() [2/2]

◆ cfgScale() [1/2]

◆ cfgScale() [2/2]

◆ params()

◆ tag()

Member Data Documentation

◆ m_desc