The paper presents a novel approach for image retrieval by combining textual and object-based visual features in order to reduce the inconsistency between the subjective user's similarity interpretation and the retrieval results produced by objective similarity models. A novel multi-scale segmentation framework is proposed to detect prominent image objects. These objects are clustered according to their visual features and mapped to related words determined by psychophysical studies. Furthermore, a hierarchy of words expressing higher-level meaning is determined on the basis of natural language processing and user evaluation. Experiments conducted on a large set of natural images showed that higher retrieval precision in terms of estimating user retrieval semantics could be achieved via this two-layer word association and also by supporting various query specifications and options.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Combining words and object-based visual features in image retrieval


    Contributors:
    Nakagawa, A. (author) / Kutics, A. (author) / Tanaka, K. (author) / Nakajima, M. (author)


    Publication date :

    2003-01-01


    Size :

    1636114 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Combining Words and Object-Based Visual Features in Image Retrieval

    Nakagawa, A. / Kutics, A. / Tanaka, K. et al. | British Library Conference Proceedings | 2003



    Combining visual dictionary, kernel-based similarity and learning strategy for image category retrieval

    Gosselin, P. H. / Cord, M. / Philipp-Foliguet, S. | British Library Online Contents | 2008


    Combining textual and visual clusters for semantic image retrieval and auto-annotation

    Celebi, E. / Alpkoca, A. | IET Digital Library Archive | 2005


    Combining Local and Global Image Features for Object Class Recognition

    Lisin, D.A. / Mattar, M.A. / Blaschko, M.B. et al. | IEEE | 2005