Content-based image retrieval (CBIR) is an attempt to remove the bottleneck of visual semantic understanding needed in automated indexing in visual information retrieval. However, the myth about the power of visual-feature-based indexing was quickly diminished as such features are far from representing semantic visual contents and producing meaningful indexes. One solution is to apply relevance feedback to refine queries or similarity measures in the search process and apply machine learning techniques to learn semantic annotations. In this paper, we address the key issues involved in relevance feedback of CBIR systems and review solutions to these issues. Based on these discussions, we present a relevance feedback and semantic learning framework for CBIR. We hope the ideas presented in this paper serve as a catalyst to more research efforts in this direction.
Learning semantics in content based image retrieval
2003-01-01
399045 byte
Conference paper
Electronic Resource
English
Learning Semantics in Content Based Image Retrieval
British Library Conference Proceedings | 2003
|Semantics Modeling Based Image Retrieval System using Neural Networks
British Library Conference Proceedings | 2005
|Probabilistic Feature Relevance Learning for Content-Based Image Retrieval
British Library Online Contents | 1999
|Image Retrieval by Content: A Machine Learning Approach
NTRS | 1994
|