Namespace EyeAuras.OpenCVAuras.ML.Yolo

Classes

YoloModelDescription

Represents a comprehensive description of a YOLO (You Only Look Once) model. This record provides a snapshot of the model's key attributes and settings, including its size, dimensions, confidence threshold, overlap threshold, output types, and a list of labels it can detect. It is used both for user information and internal processing during the inference process.

Structs

YoloClassification

Represents a classification result from a YOLO (You Only Look Once) object detection model. This struct provides details about the detected object, including its classification score, associated label, and class index.

YoloLabel

Represents a single label as identified by the YOLO (You Only Look Once) object detection system. This record structure encapsulates the unique attributes of a detected object's label, including its identifier, name, kind, and associated color.

YoloMask

Represents a binary mask used in Segmentation tasks within the Yolo model. Contains an image of the mask and the bounds of the mask.

YoloPrediction

Represents a prediction made by a Yolo model for object detection and segmentation tasks. It includes the confidence score, bounding rectangle, label, and optionally a mask for segmentation.

Interfaces

IYoloOutputParser

Defines an interface for a class responsible for parsing the outputs received from a Yolo model during the prediction process.

Enums

YoloLabelKind

Enumerates the kinds of labels used in YOLOv8 object detection. YOLO, an acronym for 'You Only Look Once', is a real-time object detection system that identifies objects in images and videos. YOLOv8 refers to the eighth version of this system. This enum distinguishes between generic labels and instance segmentation labels.

YoloModelType

Represents the different types of tasks that the Yolo model can perform.