Hi All, I am trying to create a new backend for opencv (image processing library) so that it can use directfb directly. Currently opencv for linux builds uses gtk, I am aware that there are efforts to get cairo drawing to directfb but I would like to take the more direct route. The API for opencv is very simple to extend.
My first q is, what is the prescribed way to transport arbitrary data (rgb packed for e.g.) into a Surface ? do I memcpy() ? I am playing around with the gif image provider to get a better understanding. also: I am wondering whether I should write a new provider of some kind to convert IplImage (OpenCV data structure for representing images) - using the RenderTo call or whether it would be better to extend the Blit operation so it is able to take an IplImage* as one of its arguments. The reason is I want to avoid copying each time it is to be drawn. any help would be greatly appreciated, -- :wq _______________________________________________ directfb-dev mailing list directfb-dev@directfb.org http://mail.directfb.org/cgi-bin/mailman/listinfo/directfb-dev
