Hi, I have a dataset that provides bounding boxes coordinates following this format.
height - 93 width - 108 x - 709 y- 408 label -1
Now, I want to normalize these data to train yolo.
I know there is 2 way I can normalize these values,
Way - 1 : Normalized(Xmin) = (Xmin+w/2)/Image_Width
Normalized(Ymin) = (Ymin+h/2)/Image_Height
Normalized(w) = w/Image_Width
Normalized(h) = h/Image_Height
Way - 2: divide x_center and width by image width, and y_center and height by image height.
I am not sure which way I should follow to normalize the given data? Can anyone please clear my confusion?
Hi, I have a dataset that provides bounding boxes coordinates following this format. height - 93 width - 108 x - 709 y- 408 label -1 Now, I want to normalize these data to train yolo. I know there is 2 way I can normalize these values, Way - 1 : Normalized(Xmin) = (Xmin+w/2)/Image_Width Normalized(Ymin) = (Ymin+h/2)/Image_Height Normalized(w) = w/Image_Width Normalized(h) = h/Image_Height Way - 2: divide x_center and width by image width, and y_center and height by image height.
I am not sure which way I should follow to normalize the given data? Can anyone please clear my confusion?