The proposed work exploits methods and techniques for automaticcharacterization of images for content-based access to personalphoto libraries. Several techniques, even if not reliable enough to addressthe general problem of content-based image retrieval, have been provenquite robust in a limited domain such as the one of personal photo album.In particular, starting from the observation that most personal photosdepict a usually small number of people in a relatively small numberof different contexts (e.g. Beach, Public Garden, Indoor, Nature, Snow,City, etc...) we propose the use of automatic techniques borrowed fromthe fields of computer vision and pattern recognition to index imagesbased on who is present in the scene and on the context where the picturewere taken.Experiments on a personal photo collection of about a thousand imagesproved that relatively simple content-based techniques lead to surprisinglygood results in term of easyness of user access to the data.
|Numero di pagine||10|
|Stato di pubblicazione||Published - 2007|
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- Computer Science(all)