Content Meets Semantics: Smarter Exploration of Image Collections - Presentation of Relevant Use Cases.

摘要: Current techniques for the management of image collections exploit either user-provided annotations or automatically-extracted visual features. Although effective, approach based on cannot be efficient since manual process data tagging prevents its scalability. On other hand, organization and search grounded features, such as color texture, is known to a powerful (since it can made fully automatic), yet imprecise, retrieval paradigm, because semantic gap problem. This position paper advocates combination content semantics critical binomial effectively efficiently managing browsing databases satisfying users’ expectations in quickly locating images