Intersection over Union (IoU)

It is a Quality Measure used in Image Segmentation.

Visually, it can be represented as follows.

Using the Pixels that fall in the intersection between FP, TP and FN pixel classifications, an accurate Image Segmentation can be obtained.

Mathematically

• Intersection over union ($\text{IoU}$):

$$\boxed{\mathrm{IoU}=\frac{\mathrm{TP}}{\mathrm{TP}+\mathrm{FN}+\mathrm{FP}}}$$

Here the Binary classification is directly used for calculation of IoU. Hence it is biased towards classifications of with large $\text{TP}$ and $\text{FN}$ counts.

• Instance-level intersection over union ($\text{iIoU}$):

$$\boxed{\mathrm{iIoU}=\frac{\mathrm{iTP}}{\mathrm{iTP}+\mathrm{iFN}+\mathrm{FP}}}$$

Here the $\text{TP}$ and $\text{FN}$ are weighted by the ratio of the average class instance size by the current ground truth size instance.