Overview of content-based image retrieval with high-level semantics
Semantic gap that between the visual features and human semantics has become a bottleneck of content-based image retrieval. The need for improving the retrieval accuracy of image retrieval systems and narrowing down the semantic gap is high in view of the fast growing need of image retrieval. In this paper, we first introduce the image semantic description methods, then we discuss the main technologies for reducing the semantic gap, namely, object-ontology, machine learning, relevance feedback. Applications of above-mentioned technologies in various areas are also introduced. Finally, some future research directions and problems of image retrieval are presented.
semantic mapping content-based image retriveal high-level semantics image annotation relevance feedback
Hu Min Yang Shuangyuan
Software School of Xiamen UniversityXiamen, China Software School of Xiamen University Xiamen, China
国际会议
成都
英文
1-5
2010-08-20（万方平台首次上网日期，不代表论文的发表时间）