tensorflowobject Namespace Reference

Namespace containing functions related to tensorflow object detection operations on images. More...

Classes
struct	TensorflowObject
	Represents a single depth object detection from a tensorflow model. Stores all information about a specific object detection. More...

Functions
cv::Point2f	FindObjectCenter (const TensorflowObject &stObject)
	Given an TensorflowObject struct find the center point of the corners.
std::vector< TensorflowObject >	Detect (const cv::Mat &cvFrame)
	Detect objects in the given image using a tensorflow model.

Detailed Description

Namespace containing functions related to tensorflow object detection operations on images.

Author

clayjay3 (clayt.nosp@m.onra.nosp@m.ycowe.nosp@m.n@gm.nosp@m.ail.c.nosp@m.om)

Date

2023-10-23

Function Documentation

FindObjectCenter()

cv::Point2f tensorflowobject::FindObjectCenter ( const TensorflowObject &  stObject )

inline

Given an TensorflowObject struct find the center point of the corners.

Parameters

stObject

- The object to find the center of.

Returns

cv::Point2f - The resultant center point within the image.

Author

jspencerpittman (jspen.nosp@m.cerp.nosp@m.ittma.nosp@m.n@gm.nosp@m.ail.c.nosp@m.om)

Date

2023-10-07

    {
        // Average of the four corners
        cv::Point2f cvCenter(0, 0);
 
        // Loop through each corner of the object.
        for (cv::Point2f* cvCorner : stObject.vCorners)
        {
            // Add each object x, y to the center x, y.
            cvCenter.x += cvCorner->x;
            cvCenter.y += cvCorner->y;
        }
        // Divide by number of corners.
        cvCenter.x /= 4;
        cvCenter.y /= 4;
 
        // Return a copy of the center point of the object.
        return cvCenter;
    }

Detect()

std::vector< TensorflowObject > tensorflowobject::Detect ( const cv::Mat &  cvFrame )

inline

Detect objects in the given image using a tensorflow model.

Parameters

cvFrame	- The normal BGR or BGRA image from the camera.
tfNeuralNetwork	- The tensorflow model to use for detection.

Returns

std::vector<TensorflowObject> - A vector containing all of the inferenced objects from the model.

Author

clayjay3 (clayt.nosp@m.onra.nosp@m.ycowe.nosp@m.n@gm.nosp@m.ail.c.nosp@m.om)

Date

2023-10-23

    {
        // TODO: Write different util classes to easily open and inference new models. Then put object detection specific inferencing here.
        cvFrame.empty();
        return std::vector<TensorflowObject>();
    }

Here is the call graph for this function: