bootstrap_boxes: bBootstrap bounding boxes

In dselivanov/kmeansiou: k-means clustering of with Intersection-over-Union distance

Description

bootstrap bounding boxes in order to produce smooth box distribution

Usage

bootstrap_boxes(x, scale = 2, size = nrow(x) * 10,
  quantiles = c(0.05, 0.95))

Arguments

x	input 2-d matrix: first column is wisth, second column is height of the bounding box
scale	how to rescale each box. For example 2 means it will randomly dowscaled-upscaled within a range of 1/2 to 2. Which means it's size will be within a range being 2x smaller to 2x larger then original bounding box
size	size of bootstrapped result
quantiles	keep resulting samples within these quantiles of the original distribution

dselivanov/kmeansiou documentation built on Feb. 13, 2020, 10:46 a.m.