作者: Liu Wenyin , Susan T. Dumais , Mary Czerwinski , HongJiang Zhang , Yanfeng Sun
DOI:
关键词: Quality (business) 、 Information retrieval 、 Computer science 、 User interface 、 Relevance feedback 、 Visual Word 、 Process (computing) 、 Annotation 、 Image retrieval 、 Automatic image annotation
摘要: A novel approach to semi-automatically and progressively annotating images with keywords is presented. The progressive annotation process embedded in the course of integrated keyword-based content-based image retrieval user feedback. When submits a keyword query then provides relevance feedback, search are automatically added that receive positive feedback can facilitate future. coverage quality such database system improved as cycle increases. strategy semi-automatic better than manual terms efficiency automatic accuracy. performance study presented which shows high be achieved this approach, preliminary described showing users view annotations important will likely use them retrieval. also suggested interface enhancements needed support We believe similar approaches could applied managing other forms multimedia objects.