InstanceSegmentationBase<T>

Base Classes · AiDotNet.ComputerVision.Segmentation.Common

Abstract base class for instance segmentation models that detect and mask individual object instances.

Inheritance: NeuralNetworkBase<T> → SegmentationModelBase<T> → InstanceSegmentationBase<T>

Implements: IInstanceSegmentation<T>

Known derived types: YOLO11Seg<T>, YOLO26Seg<T>, YOLOv12Seg<T>, YOLOv8Seg<T>, YOLOv9Seg<T>

For Beginners

Instance segmentation finds each individual object in an image and provides a pixel-level mask for each one. Unlike semantic segmentation (all cars are “car”), instance segmentation distinguishes car #1 from car #2.

Models extending this base class: YOLOv9-Seg, YOLO11-Seg, YOLOv12-Seg, YOLO26-Seg, Mask2Former, MaskDINO.

Properties

Property Summary
double ConfidenceThreshold { get; set; }
double NmsThreshold { get; set; }
int MaxInstances { get; }

Methods

Method Summary
InstanceSegmentationResult<T> DetectInstances(Tensor<T> image)