I would have thought that adobe would have either invented or bought patents relating to context aware fill in order to prevent other graphics software from implementing it. I am very pleased to see it in Pixelmator.
Read his original PhD thesis on "PatchMatch: A Randomized Correspondence Algorithm for Structural Image Editing" http://gfx.cs.princeton.edu/pubs/Barnes_2009_PAR/index.php
As well as the follow up paper with collaboration from Adobe: "Video Tapestries with Continuous Temporal Zoom" (published at SIGGRAPH) http://gfx.cs.princeton.edu/pubs/Barnes_2010_VTW/index.php