Package org.photonvision.vision.objects

Interface ObjectDetector

All Superinterfaces:
Releasable
All Known Implementing Classes:
NullModel, RknnObjectDetector
public interface ObjectDetector extends Releasable
ObjectDetector lifecycle:
  1. Models are discovered by NeuralNetworkModelManager
  2. Model is selected as a parameter of ObjectDetectionPipe
  3. Model.load() is called to create a ObjectDetector instance
  4. detect(Mat, double, double) is called to perform object detection
  5. Releasable.release() is called to release resources

  • Method Summary

    Modifier and Type
    Method
    Description
    List<NeuralNetworkPipeResult>
    detect(org.opencv.core.Mat in, double nmsThresh, double boxThresh)
    Detects objects in the given input image.
    List<String>
    getClasses()
    Returns the classes that the detector can detect
    Model
    getModel()
    Returns the model that created this ObjectDetector.

    Methods inherited from interface org.photonvision.vision.opencv.Releasable

    release

  • Method Details

    • getModel

      Model getModel()
      Returns the model that created this ObjectDetector.

    • getClasses

      List<String> getClasses()
      Returns the classes that the detector can detect
      Returns:
      The classes

    • detect

      List<NeuralNetworkPipeResult> detect(org.opencv.core.Mat in, double nmsThresh, double boxThresh)
      Detects objects in the given input image. Preprocessing and postprocessing steps should be embedded into this call.
      Parameters:
      in - The input image to perform object detection on.
      nmsThresh - The threshold value for non-maximum suppression.
      boxThresh - The threshold value for bounding box detection.
      Returns:
      A list of NeuralNetworkPipeResult objects representing the detected objects. Returns an empty list if the detector is not initialized or if no objects are detected.