An Open Framework for Distributed Multimedia Retrieval

Bibtex entry :

@inproceedings { VG:MMS2000c,
    author = { Henning M{\"u}ller and Wolfgang M{\"u}ller and David McG. Squire and Zoran Pe\u{c}enovi\'{c} and St{\'e}phane Marchand-Maillet and Thierry Pun },
    title = { An Open Framework for Distributed Multimedia Retrieval },
    booktitle = { Recherche d'Informations Assist\'ee par Ordinateur (RIAO'2000) Computer-AssistedInformation Retrieval },
    pages = { 701--712. },
    year = { 2000 },
    volume = { 1 },
    address = { Paris, France },
    month = { apr 12-14 },
    url = { },
    abstract = { This article describes a framework for distributed multimedia retrievalwhich permits the connection of compliant user interfaces with avariety of multimedia retrieval engines via an open communicationprotocol, MRML (Multi Media Retrieval Markup Language). It allowsthe choice of image collection, feature set and query algorithm duringrun--time, permitting multiple users to query a system adapted totheir needs, using the query paradigm adapted to their problem suchas query by example (QBE), browsing queries, or query by annotation.User interaction is implemented over several levels and in diverseways. Relevance feedback is implemented using positive and negativeexample images that can be used for a best--match QBE query. In contrast,browsing methods try to approach the searched image by giving overviewsof the entire collection and by successive refinement. In additionto these query methods, Long term off line learning is implemented.It allows feature preferences per user, user domain or over all usersto be learned automatically. We present the Viper multimedia retrievalsystem as the core of the framework and an example of an MRML-compliantsearch engine. Viper uses techniques adapted from traditional informationretrieval (IR) to retrieve multimedia documents, thus benefitingfrom the many years of IR research. As a result, textual and visualfeatures are treated in the same way, facilitating true multimediaretrieval. The MRML protocol also allows other applications to makeuse of the search engnes. This can for example be used for the designof a benchmark test suite, querying several search engines in thesame way and comparing the results. This is motivated by the factthat the content--based image retrieval community really lacks sucha benchmark as it already exists in text retrieval. },
    owner = { steph },
    timestamp = { 2008.05.04 },
    url1 = { },
    vgclass = { refpap },
    vgproject = { viper },

Keywords: machine learning, information geometry, data mining, Big Data, affective information retrieval (recherche d'information), information visualisation, content-based image and video retrieval (CBIR, CBR, CBVR, CBMR, CBMIR), information mining, classification, multimedia and multimodal information management, semantic web, knowledge base (RDF, OWL, XML, metadata, auto-annotation, description), multimodal information fusion