Revision history [back]

I'm working on face detection. I'm implementing the Viola-Jones algorithm on an FPGA device. I'm using a pre-trained classifier but I can't understand how Opencv handles rectangles feature. For example, suppose that subwindow size is 24x24 and a rectangle has x = 0, y = 0, w = 24, h = 3. Actually I know that rectangles are handled like this x =< pt.x < x+w y =< pt.y < y+h, but it's a non-sense because x = 0, y = 0, w = 8, h = 1 isn't a rectangle, it's a line. So I hope someone can help me. Thanks