Interface: InstanceSegmentationOptions<L>

Defined in: types/instanceSegmentation.ts:44

Options for instance segmentation forward pass.

Type Parameters

L

L extends LabelEnum

The label map type for the model, must conform to LabelEnum.

Properties

classesOfInterest?

optional classesOfInterest: keyof L[]

Defined in: types/instanceSegmentation.ts:62

Filter to include only specific classes.

---

confidenceThreshold?

optional confidenceThreshold: number

Defined in: types/instanceSegmentation.ts:49

Minimum confidence threshold for including instances. Defaults to model's defaultConfidenceThreshold (typically 0.5).

---

inputSize?

optional inputSize: number

Defined in: types/instanceSegmentation.ts:72

Input size for the model (e.g., 384, 512, 640). Must be one of the model's availableInputSizes. Defaults to model's defaultInputSize.

---

iouThreshold?

optional iouThreshold: number

Defined in: types/instanceSegmentation.ts:54

IoU threshold for non-maximum suppression. Defaults to model's defaultIouThreshold (typically 0.5).

---

maxInstances?

optional maxInstances: number

Defined in: types/instanceSegmentation.ts:58

Maximum number of instances to return. Default: 100

---

returnMaskAtOriginalResolution?

optional returnMaskAtOriginalResolution: boolean

Defined in: types/instanceSegmentation.ts:66

Whether to return masks at original image resolution. Default: true