A New Gap to Bridge: Where to Go Next in Social Media Retrieval?

(Extended Abstract)
  • Alan Hanjalic
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7131)


Research in Multimedia Information Retrieval (MIR) aims at matching multimedia content and user needs and so at bringing image, audio and video content together with users. Users expectations regarding multimedia content access in terms of semantically rich and personalized relevance criteria have always been high and have imposed high demands on the level of sophistication of MIR solutions. The potential to develop MIR technology that meets such high demands has rapidly grown over the past twenty years by building on intensive international research effort. This growth accelerated, however, with the increasing contextualization of images, video and music in rapidly expanding social networks that link distributed content, diverse metadata and users of various profiles and interests. It is clear that user demands regarding the sophistication of MIR technology have further grown in the social network context in view of new ways of interacting with multimedia content and with other people via and about this content. However, this new context has also brought vast new opportunities for improving the quality of MIR solutions. These opportunities lie in synergetic integrations of multidisciplinary scientific contributions and rich information resources found there. Revisiting MIR from the viewpoint of the social network context, using the approaches that are often jointly referred to as social media retrieval, can help the field not only resolve the problems that impeded its development in the past, but also address the new emerging demands. I will show how contextualizing the MIR in online networked communities of users can help us achieve a fundamental shift in the MIR grand challenge, from bridging the research-oriented semantic gap to bridging the much more important, user-oriented utility gap, that explicitly addresses the overall usefulness of a MIR system output for the user. I will highlight some of the opportunities in pursuing this new, utility-oriented MIR grand challenge.


Social Network Computer Vision High Demand International Research Extend Abstract 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Alan Hanjalic
    • 1
  1. 1.Delft Multimedia Information Retrieval LabDelft University of TechnologyDelftThe Netherlands

Personalised recommendations