0

I'm currently working on a project using OpenCV and dlib. I wanted to convert the bounding box from the dlib detector to the bounding box format in OpenCV.

I looked it up and found this link: Boundingbox defintion for opencv object tracking

It says that OpenCV follows the format (x, y, w, h) where x and y are the xmin and ymin. Makes sense. But it also says that (xmin, ymin) is the top left point of the bounding box. How is that possible? Shouldn't the top left point be (xmin, ymax)?

Anubhav Guha
  • 1
  • 3
  • Images are indexed from top to bottom in OpenCV, thus top left corner is `(xmin, ymin)`. – HansHirse Jun 01 '21 at 08:49
  • Okay, so top left from the perspective of the image but (xmin, ymin) from the perspective of the origin of the image. Got it, thanks! – Anubhav Guha Jun 01 '21 at 09:01

0 Answers0