We present a Bayesian learning algorithm that relies on belief propagation to integrate feedback provided by the user over a retrieval session. Bayesian retrieval leads to a natural criteria for evaluating local image similarity without requiring any image segmentation. This allows the practical implementation of retrieval systems where users can provide image regions, or objects, as queries. Region-based queries are significantly less ambiguous than queries based on entire images leading to significant improvements in retrieval precision. When combined with local similarity, Bayesian belief propagation is a powerful paradigm for user interaction. Experimental results show that significant improvements in the frequency of convergence to the relevant images can be achieved by the inclusion of learning in the retrieval process.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Bayesian relevance feedback for content-based image retrieval


    Contributors:
    Vasconcelos, (author) / Lippman, (author)


    Publication date :

    2000-01-01


    Size :

    184176 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Bayesian Relevance Feedback for Content-based Image Retrieval

    Vasconcelos, N. / Lippman, A. / IEEE | British Library Conference Proceedings | 2000


    Multi-class relevance feedback content-based image retrieval

    Peng, J. | British Library Online Contents | 2003


    Interactive Content-Based Image Retrieval Using Relevance Feedback

    MacArthur, S. D. / Brodley, C. E. / Kak, A. C. et al. | British Library Online Contents | 2002


    Relevance feedback decision trees in content-based image retrieval

    MacArthur, / Brodley, / Chi-Ren Shyu, | IEEE | 2000


    Relevance Feedback Decision Trees in Content-based Image Retrieval

    MacArthur, S. / Brodley, C. / Shyu, C. et al. | British Library Conference Proceedings | 2000