0
$\begingroup$

For the image datasets, there may be a bounding box for each image at the dataset. It is an annotation for an image. It is a rectangular box intended for focusing on something inside the image.

I read about the following two types of representations for a bounding box.

  1. using two points $(x_1, y_1)$ and $(x_2, y_2)$.

$$<x_1><y_1><x_2><y_2>$$

  1. Using a point $(x_1, y_1)$, width, and height. $$<x_1><y_1><width><height>$$

How do understand both the representations, Specifically, does the point $(x_1, y_1)$ used to denote the top right or top left or bottom right or bottom left in both cases?

$\endgroup$
1
  • 1
    $\begingroup$ In case 2 it could also be the centre, ie the box starts at (x - (w/2)) and end at (x+(w/2)) etc. $\endgroup$ Sep 20 at 10:39
0
$\begingroup$

With Bounding box people usually mean literally the box which bounds something. So, more intuitive way to use the points in the first case as bottom left and top right points and in the second case bottom left point and sizes. This corresponds to the logic of points on the Cartesian grid. Of course, you may reverse sometimes against Y axis in case your task requires so. For example, some software has X origin at top left corner of the monitor. In this case, sometimes y subscripts can reverse.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.