Base class for neural network classifiers. More...

Inheritance diagram for core::NeuralClassifierBase:

Collaboration diagram for core::NeuralClassifierBase:

Public Member Functions
	NeuralClassifierBase ()=default
	Default constructor.

std::map< std::string, double >	runNetwork (const cv::Mat &frame, core::Detection &detection)
	Run the classification network on the given frame. More...

Public Member Functions inherited from core::NeuralBase
std::string	getName ()
	Gets the name of the neural network. More...

double	getInferenceTime ()
	Gets the inference time of the network. More...

void	loadNetwork (std::string framework, std::string modelWeightsFile, std::string classesFile="", std::string modelConfigurationFile="", std::string inputLayerName="", std::string outputLayerName="")
	Loads the neural network from memory buffer. More...

void	configureNetwork (std::string name, int inputWidth, int inputHeight, double scale, double mean0, double mean1, double mean2, std::string configBackend, std::string configTarget, double confidenceThreshold)
	Configures the network with various parameters. More...

std::vector< std::string >	getOutputsNames (const cv::dnn::Net &net)
	Gets the names of the output layers of the network. More...

Additional Inherited Members
Protected Member Functions inherited from core::NeuralBase
void	readClassNames (const std::string &filePath)

Protected Attributes inherited from core::NeuralBase
std::string	mFramework

std::string	mInputLayerName

std::string	mOutputLayerName

std::string	mName

cv::dnn::Net	mNet

std::vector< std::string >	classes

double	scaleFactor = 1.0

cv::Scalar	mean { 0.0, 0.0, 0.0 }

cv::Size	size { 128, 128 }

cv::dnn::Backend	selectedBackend

cv::dnn::Target	selectedTarget

float	confThreshold = 0.5F

std::map< std::string, cv::dnn::Backend >	backends

std::map< std::string, cv::dnn::Target >	targets

Detailed Description

Base class for neural network classifiers.

Provides a convenience wrapper to execute a neural network and obtain a map of class probabilities.

Member Function Documentation

std::map< std::string, double > core::NeuralClassifierBase::runNetwork	(	const cv::Mat &	frame,
		core::Detection &	detection
	)

Run the classification network on the given frame.

Parameters

frame	Image to classify.
detection	Detection information associated with the image.

Here is the caller graph for this function:

The documentation for this class was generated from the following files: