On May 2, 2012, at 5:36 PM, Jeremy Selan wrote: > > - Your proposal give lots of examples that require complex interfaces (such > as seamless cloning). Our initial thoughts are that we'd prefer for you to > focus on the core API for the solver/image transform operators, and to > totally ignore interface. For example, in the seamless cloning example, > we'll want you to assume that an alpha image representing the area to clone > already exists. You should not worry about adding roto/boundary selection to > OIIO.
Right. For now write the ImageBufAlgo functions accepting a matte image (in addition to the other inputs). Save UI or other issues about how to specify or derive the matte image until later, and we may ultimately decide that kind of thing is way outside the scope of OIIO proper. -- Larry Gritz l...@larrygritz.com _______________________________________________ Oiio-dev mailing list Oiio-dev@lists.openimageio.org http://lists.openimageio.org/listinfo.cgi/oiio-dev-openimageio.org
