نبذة مختصرة : We propose a new system for editing personal photo collections, inspired by search-and-replace editing for text. In our system, local edits specified by the user in a single photo (e.g., using the “clone brush” tool) can be propagated automatically to other photos in the same collection, by matching the edited region across photos. To achieve this, we build on tools from computer vision for image matching. Our experimental results on real photo collections demonstrate the feasibility and potential benefits of our approach. ; Natural Sciences and Engineering Research Council of Canada Postdoctoral Fellowship ; Massachusetts Institute of Technology. Undergraduate Research Opportunities Program ; National Science Foundation (U.S.) (CAREER award 0447561) ; T-Party Project ; United States. National Geospatial-Intelligence Agency (NGA NEGI-1582- 04-0004) ; United States. Office of Naval Research. Multidisciplinary University Research Initiative (Grant N00014-06-1-0734) ; Microsoft Research ; Alfred P. Sloan Foundation
No Comments.