This paper describes our efforts to develop a Digital Office in which we augment a physical office setting with cameras and other electronic devices. Our goal is to bring the worlds of electronic and physical documents closer together and to facilitate the interaction of humans with all kinds of documents. In the Digital Office we extend the traditional notion of scanning documents to include the capture of whiteboards, books, desktops, and the human office workers themselves. In particular, we give an overview of three systems in which video cameras unobtrusively observe and capture whiteboards and human gestures, papers on the desktop, and the motion of a user’s face which is used to control the display of electronic documents in a browser. Each of these systems combines the features and affordances of both physical and electronic documents, and together they begin to illuminate the intelligent office environment of the future.