In this function,we get the
orig_size = np.array([images[0].width, images[0].height])
and then
boxes_xy = [boxxy / orig_size for boxxy in boxes_xy]
boxes_wh = [boxwh / orig_size for boxwh in boxes_wh]
We get only an image size . Shouldn't we divide the boxes_xy or boxes_wh by their corresponding image size ?
Is it a bug or im stupid?
sorry for my bad English. i hope you get it.
In this function,we get the orig_size = np.array([images[0].width, images[0].height]) and then boxes_xy = [boxxy / orig_size for boxxy in boxes_xy] boxes_wh = [boxwh / orig_size for boxwh in boxes_wh]
We get only an image size . Shouldn't we divide the boxes_xy or boxes_wh by their corresponding image size ? Is it a bug or im stupid? sorry for my bad English. i hope you get it.