An Approach to Semantics-based Image Retrieval and Browsing

The state-of-the-art image retrieval approach is to incorporate image semantics with low-level features to enhance retrieval performance. Although many systems annotate images with descriptive keywords and retrieve images by keyword-based search, they have not explored the full potentials of semantics. This paper exploits the power of thesaurus-aided approaches to facilitate semantics-based access to image collections. The contribution of our work includes constructing a dynamic semantic hierarchy (DSH) to support flexible browsing of images by semantic subjects, as well as formulating a semantic similarity metric to improve the accuracy of semantic matching. Both approaches are seamlessly integrated into a unified framework for semanticsand feature-based image retrieval. Experiments conducted on real-world images demonstrate the effectiveness of our approaches.