MediaPipeTasksVision Framework Reference

ImageSegmenterResult

class ImageSegmenterResult : TaskResult

Represents the segmentation results generated by ImageSegmenter.

  • An optional array of Mask objects. Each Mask in the array holds a 32 bit float array of size image width * image height which represents the confidence mask for each category. Each element of the float array represents the confidence with which the model predicted that the corresponding pixel belongs to the category that the mask represents, usually in the range [0,1].

    Declaration

    Swift

    var confidenceMasks: [Mask]? { get }

  • An optional Mask that holds aUInt8 array of size image width * image height. Each element of this array represents the class to which the pixel in the original image was predicted to belong to.

    Declaration

    Swift

    var categoryMask: Mask? { get }

  • The quality scores of the result masks, in the range of [0, 1]. Defaults to 1 if the model doesn’t output quality scores. Each element corresponds to the score of the category in the model outputs.

    Declaration

    Swift

    var qualityScores: [NSNumber]? { get }

  • Initializes a new ImageSegmenterResult with the given array of confidence masks, category mask, quality scores and timestamp (in milliseconds).

    Declaration

    Swift

    init(confidenceMasks: [Mask]?, categoryMask: Mask?, qualityScores: [NSNumber]?, timestampInMilliseconds: Int)

    Parameters

    confidenceMasks

    An optional array of Mask objects. Each Mask in the array must be of type float32.
    categoryMask

    An optional Mask object of type uInt8.
    qualityScores

    The quality scores of the result masks of type NSArray *. Each NSNumber in the array holds a float.
    timestampInMilliseconds

    The timestamp (in milliseconds) for this result.

    Return Value

    An instance of ImageSegmenterResult initialized with the given array of confidence masks, category mask, quality scores and timestamp (in milliseconds).