Implicit Visual Concept Modeling in Image / Video Annotation

摘要

In this paper a novel approach for automatically annotating image databases is proposed. Despite most current approaches that are just based on spatial content analysis, the proposed method properly combines implicit feedback information and visual concept models for semantically annotating images. Our method can be easily adopted by any multimedia search engine, providing an intelligent way to even annotate completely non-annotated content. The proposed approach currently provides very interesting results in limited-content environments and it is expected to add significant value to billions of non-annotated images existing in the Web. Furthermore expert annotators can gain important knowledge relevant to user new trends, language idioms and styles of searching.