Lihi Zelnik and Pietro Perona
A single picture cannot always capture the experience of 'being there'.
When we explore a complex scene we move about and view it from a number of different vantage points.
Artists have long known that they can violate the rules of combine different viewpoints into a single painting for better effect.
Recently, David Hockney put together a number of 'joiners' obtained gluing together a number of photographs taken from different viewpoints and at different times. In a typical joiner some images occlude others and boundaries across pictures are fully visible. In spite (or maybe due to) their fragmented look, such representations are interesting and effective. Hockney's joiners have inspired a number of photographers and artists (e.g., on Flickr.com search for tags 'joiner', 'Hockney', 'photocollage'). We are interested in automating the construction of such image joiners for two main objectives. The first is providing photographers with a good starting point from which they can more easily finalize their joiners. The second is providing a new way for exploring image collections now organized in correspondence with the underlying scene rather than by file names.
(For more information click here.)