digiKam
Public Member Functions | Static Public Attributes | Protected Member Functions | Protected Attributes | List of all members
Digikam::DNNFaceDetectorSSD Class Reference
+ Inheritance diagram for Digikam::DNNFaceDetectorSSD:

Public Member Functions

void detectFaces (const cv::Mat &inputImage, const cv::Size &paddedSize, std::vector< cv::Rect > &detectedBboxes) override
 
 DNNFaceDetectorSSD ()
 
bool loadModels ()
 
cv::Size nnInputSizeRequired () const
 
 ~DNNFaceDetectorSSD () override
 

Static Public Attributes

static float confidenceThreshold = 0.6F
 Threshold for bbox detection. It can be init and changed in the GUI. More...
 
static float nmsThreshold = 0.4F
 Threshold for nms suppression. More...
 

Protected Member Functions

void correctBbox (cv::Rect &bbox, const cv::Size &paddedSize) const
 
void selectBbox (const cv::Size &paddedSize, float confidence, int left, int right, int top, int bottom, std::vector< float > &goodConfidences, std::vector< cv::Rect > &goodBoxes, std::vector< float > &doubtConfidences, std::vector< cv::Rect > &doubtBoxes) const
 

Protected Attributes

cv::Size inputImageSize
 
cv::Scalar meanValToSubtract
 
QMutex mutex
 
cv::dnn::Net net
 
float scaleFactor
 

Constructor & Destructor Documentation

◆ DNNFaceDetectorSSD()

Digikam::DNNFaceDetectorSSD::DNNFaceDetectorSSD ( )
explicit

References loadModels().

◆ ~DNNFaceDetectorSSD()

Digikam::DNNFaceDetectorSSD::~DNNFaceDetectorSSD ( )
override

Member Function Documentation

◆ correctBbox()

void Digikam::DNNFaceDetectorBase::correctBbox ( cv::Rect &  bbox,
const cv::Size &  paddedSize 
) const
protectedinherited

References Digikam::DNNFaceDetectorBase::inputImageSize.

◆ detectFaces()

void Digikam::DNNFaceDetectorSSD::detectFaces ( const cv::Mat &  inputImage,
const cv::Size &  paddedSize,
std::vector< cv::Rect > &  detectedBboxes 
)
overridevirtual

Implements Digikam::DNNFaceDetectorBase.

References Digikam::DNNFaceDetectorBase::inputImageSize, Digikam::DNNFaceDetectorBase::meanValToSubtract, Digikam::DNNFaceDetectorBase::mutex, Digikam::DNNFaceDetectorBase::net, and Digikam::DNNFaceDetectorBase::scaleFactor.

◆ loadModels()

bool Digikam::DNNFaceDetectorSSD::loadModels ( )

References Digikam::DNNFaceDetectorBase::net.

Referenced by DNNFaceDetectorSSD().

◆ nnInputSizeRequired()

cv::Size Digikam::DNNFaceDetectorBase::nnInputSizeRequired ( ) const
inherited

References Digikam::DNNFaceDetectorBase::inputImageSize.

◆ selectBbox()

void Digikam::DNNFaceDetectorBase::selectBbox ( const cv::Size &  paddedSize,
float  confidence,
int  left,
int  right,
int  top,
int  bottom,
std::vector< float > &  goodConfidences,
std::vector< cv::Rect > &  goodBoxes,
std::vector< float > &  doubtConfidences,
std::vector< cv::Rect > &  doubtBoxes 
) const
protectedinherited

Classify bounding boxes detected. Good bounding boxes are defined as boxes that reside within the non-padded zone or those that are out only for min of (10% of padded range, 10% of bbox dim).

Bad bounding boxes are defined as boxes that have at maximum 25% of each dimension out of non-padded zone.

References Digikam::DNNFaceDetectorBase::inputImageSize.

Member Data Documentation

◆ confidenceThreshold

float Digikam::DNNFaceDetectorBase::confidenceThreshold = 0.6F
staticinherited

Threshold for bbox detection. It can be init and changed in the GUI.

◆ inputImageSize

cv::Size Digikam::DNNFaceDetectorBase::inputImageSize
protectedinherited

Referenced by Digikam::DNNFaceDetectorBase::correctBbox(), detectFaces(), Digikam::DNNFaceDetectorYOLO::detectFaces(), Digikam::DNNFaceDetectorBase::nnInputSizeRequired(), and Digikam::DNNFaceDetectorBase::selectBbox().

◆ meanValToSubtract

cv::Scalar Digikam::DNNFaceDetectorBase::meanValToSubtract
protectedinherited

Referenced by detectFaces(), and Digikam::DNNFaceDetectorYOLO::detectFaces().

◆ mutex

QMutex Digikam::DNNFaceDetectorBase::mutex
protectedinherited

Referenced by detectFaces(), and Digikam::DNNFaceDetectorYOLO::detectFaces().

◆ net

cv::dnn::Net Digikam::DNNFaceDetectorBase::net
protectedinherited

Referenced by detectFaces(), Digikam::DNNFaceDetectorYOLO::detectFaces(), loadModels(), and Digikam::DNNFaceDetectorYOLO::loadModels().

◆ nmsThreshold

float Digikam::DNNFaceDetectorBase::nmsThreshold = 0.4F
staticinherited

Threshold for nms suppression.

◆ scaleFactor

float Digikam::DNNFaceDetectorBase::scaleFactor
protectedinherited

Referenced by detectFaces(), and Digikam::DNNFaceDetectorYOLO::detectFaces().

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