MLObject

Datatype for storing detected objects for Object Detector. An MLObject contains a String label, confidence score, and bounding box information (width, height, and the x and y coordinates of the upper left corner).

Examples

void setup() {
    // create an Object Detector
    ObjectDetector detector = new ObjectDetector(this, "coco_ssd");
    // load input image
    PImage img = loadImage("data/dog_bike_car.jpeg");
    // run the object detection
    MLObject[] output = detector.predict(img);
    // print image label and confidence of each object
    for (int i = 0; i < output.length; i++) {
        String label = output[i].getLabel();
        float confidence = output[i].getConfidence();
        println(label + " detected! (confidence: " + confidence + ")");
    }
}

Methods

• getLabel(): Returns String label that stores the label name of the object.
• getConfidence(): Returns float confidence that stores the confidence score (from 0 to 1) of the object. As the confidence score is higher, the possibility of that object is labelled correclty is higher.
• getX(): Returns float x, the x coordinate of the upper left corner of the bounding box.
• getY(): Returns float y, the y coordinate of the upper left corner of the bounding box.
• getWidth(): Returns float width, the width of the bounding box.
• getHeight(): Returns float height, the height of the bounding box.

Used By

Object Detector