cv::dnn_objdetect::InferBbox Class Reference

A class to post process model predictions. More...

#include <opencv2/core_detect.hpp>

Collaboration diagram for cv::dnn_objdetect::InferBbox:

Public Member Functions
	InferBbox (Mat _delta_bbox, Mat _class_scores, Mat _conf_scores)
	Default constructer.
void	filter (double thresh=0.8)
	Filters the bounding boxes.

Public Attributes
std::vector< object >	detections
	Vector which holds the final detections of the model.

Protected Member Functions
void	assert_predictions (std::vector< std::vector< double > > *min_max_boxes)
	Ensures that the bounding box values are within image boundaries.
void	filter_top_n (std::vector< std::vector< double > > *probs, std::vector< std::vector< double > > *boxes, std::vector< std::vector< double > > &top_n_boxes, std::vector< size_t > &top_n_idxs, std::vector< double > &top_n_probs)
	Filter top n predictions.
void	final_probability_dist (std::vector< std::vector< double > > *final_probs)
	Computes final probability values of each bounding box.
void	intersection_over_union (std::vector< std::vector< double > > *boxes, std::vector< double > *base_box, std::vector< double > *iou)
	Computes intersection over union of bounding boxes.
void	nms_wrapper (std::vector< std::vector< double > > &top_n_boxes, std::vector< size_t > &top_n_idxs, std::vector< double > &top_n_probs)
	Wrapper to apply Non-Maximal Supression.
std::vector< bool >	non_maximal_suppression (std::vector< std::vector< double > > *boxes, std::vector< double > *probs)
	Applies Non-Maximal Supression.
void	transform_bboxes (std::vector< std::vector< double > > *bboxes)
	Transform relative coordinates from ConvDet to bounding box coordinates.
void	transform_bboxes_inv (std::vector< std::vector< double > > *pre, std::vector< std::vector< double > > *post)
	Transform bounding boxes from [x, y, h, w] to [xmin, ymin, xmax, ymax].

Static Protected Member Functions
static bool	comparator (std::pair< double, size_t > l1, std::pair< double, size_t > l2)

Detailed Description

A class to post process model predictions.

Constructor & Destructor Documentation

InferBbox()

cv::dnn_objdetect::InferBbox::InferBbox	(	Mat	_delta_bbox,
		Mat	_class_scores,
		Mat	_conf_scores
	)

Default constructer.

Parameters

_delta_bbox	Blob containing relative coordinates of bounding boxes
_class_scores	Blob containing the probability values of each class
_conf_scores	Blob containing the confidence scores

Member Function Documentation

assert_predictions()

void cv::dnn_objdetect::InferBbox::assert_predictions ( std::vector< std::vector< double > > *  min_max_boxes )

protected

Ensures that the bounding box values are within image boundaries.

Parameters

min_max_boxes

Vector containing bounding boxes of the form [xmin, ymin, xmax, ymax]

comparator()

static bool cv::dnn_objdetect::InferBbox::comparator ( std::pair< double, size_t >  l1, std::pair< double, size_t >  l2  )

inlinestaticprotected

filter()

void cv::dnn_objdetect::InferBbox::filter

(

double

thresh = 0.8

)

Filters the bounding boxes.

filter_top_n()

void cv::dnn_objdetect::InferBbox::filter_top_n ( std::vector< std::vector< double > > *  probs, std::vector< std::vector< double > > *  boxes, std::vector< std::vector< double > > &  top_n_boxes, std::vector< size_t > &  top_n_idxs, std::vector< double > &  top_n_probs  )

protected

Filter top n predictions.

Parameters

probs	Final probability values of bounding boxes
boxes	Predicted bounding box co-ordinates
top_n_boxes	Contains bounding box co-ordinates of top n boxes
top_n_idxs	Containes class indices of top n bounding boxes
top_n_probs	Contains probability values of top n bounding boxes

final_probability_dist()

void cv::dnn_objdetect::InferBbox::final_probability_dist ( std::vector< std::vector< double > > *  final_probs )

protected

Computes final probability values of each bounding box.

Parameters

final_probs

Vector to hold the probability values

intersection_over_union()

void cv::dnn_objdetect::InferBbox::intersection_over_union ( std::vector< std::vector< double > > *  boxes, std::vector< double > *  base_box, std::vector< double > *  iou  )

protected

Computes intersection over union of bounding boxes.

Parameters

boxes	Vector of bounding box co-ordinates
base_box	Base box wrt which IOU is calculated
iou	Vector to store IOU values

nms_wrapper()

void cv::dnn_objdetect::InferBbox::nms_wrapper ( std::vector< std::vector< double > > &  top_n_boxes, std::vector< size_t > &  top_n_idxs, std::vector< double > &  top_n_probs  )

protected

Wrapper to apply Non-Maximal Supression.

Parameters

top_n_boxes	Contains bounding box co-ordinates of top n boxes
top_n_idxs	Containes class indices of top n bounding boxes
top_n_probs	Contains probability values of top n bounding boxes

non_maximal_suppression()

std::vector< bool > cv::dnn_objdetect::InferBbox::non_maximal_suppression ( std::vector< std::vector< double > > *  boxes, std::vector< double > *  probs  )

protected

Applies Non-Maximal Supression.

Parameters

boxes	Bounding box co-ordinates belonging to one class
probs	Probability values of boxes belonging to one class

transform_bboxes()

void cv::dnn_objdetect::InferBbox::transform_bboxes ( std::vector< std::vector< double > > *  bboxes )

protected

Transform relative coordinates from ConvDet to bounding box coordinates.

Parameters

bboxes

Vector to hold the predicted bounding boxes

transform_bboxes_inv()

void cv::dnn_objdetect::InferBbox::transform_bboxes_inv ( std::vector< std::vector< double > > *  pre, std::vector< std::vector< double > > *  post  )

protected

Transform bounding boxes from [x, y, h, w] to [xmin, ymin, xmax, ymax].

Parameters

pre	Vector conatining initial co-ordinates
post	Vector containing the transformed co-ordinates

Member Data Documentation

detections

std::vector<object> cv::dnn_objdetect::InferBbox::detections

Vector which holds the final detections of the model.

---

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

• opencv2/core_detect.hpp