Hi,
Let’s assume that I run an inference in an image with dimensions 1000 x 2000px with Yolo v9 in order to detect some objets.
When running this analysis with ‘imgsz = 1000’ I will get the following message :
imgsz=[1000] must be multiple of max stride 32, updating to [xxxxx]
So I guess that in the background the image dimensions should be a multiple of 32. Ok, no problems.
My question is : the resulted bounding boxes coordinates are from the original image or the resized one ?
Thank you all !
Hi Igor, how is it going?
I think this answer may help you.
Closing this post following the below answer.
system
Closed
4
This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.
