Why are bounding box coordinates in decimal points ?

[Leothorn]

I have been trying to understand what the print(outputs["instances"].pred_boxes) represents. I now know that they represent the coordinates of the boxes detected. But why are these coordinates in decimals (or float values)? Why are they not whole numbers?

Normally we would have coordinates starting at (0,0) in the top left corner of the image and the next pixel would be (0,1) in (x,y) format. But, as shown by @deeplearner93 here, we obtain values like (126.6035,244.8977). Why is this the case?

@ sushmasuresh28