Sites Inria

Version française

LINKMEDIA Research team

Creating and exploiting explicit links between multimedia fragments

Team presentation

The challenge that multimedia faces today is that of context awareness, i.e., describing documents in the context in which they appear (context of a collection, social context, etc.). Following this line of thought, the seminal idea of LinkMedia is that of content-based media linking with the ultimate goal of enabling better multimedia applications and new innovative services. Taking a content-based perspective, we seek to create explicit links at different levels to better reflect the context: links at the signal level, e.g., with repeating patterns; links at a semantic level, e.g., to follow topics or stories; links at a paradigmatic level, e.g., to have further details or comments on a topic. LinkMedia investigates a number of key issues related to multimedia collections with explicit links: Can we discover what characterizes a collection and makes its coherence? Are there repeating motifs that create natural links and which deserve characterization and semantic interpretation? How to explicitly create links from pairwise distances? What structure should a linked collection have? How do we explain the semantic of a link? How explicit links can be used to improve information retrieval? To improve user experience? Addressing such questions, our goal is to lay down scientific foundations for collection structuring by means of explicit links and to study new usages and content processing techniques induced by structured context-aware collections.

Research themes

Pursuing the global objective of linked media collections, LinkMedia seeks to develop methodological, technological and scientific foundations to create, describe and exploit explicit links within multimedia collections on a large scale. We instantiate research activities along three major lines of research, namely 1. unsupervised motif discovery ? improve multimedia indexing for efficient comparison of content items, develop scalable algorithms for the unsupervised discovery of repeated motifs, investigate motif interpretation and exploitation 2. description and structuring of multimedia content ? improve technology for multimedia content modeling and for semantic interpretation, offering a better trade-off between accuracy, robustness and scalability 3. linking of multimedia content ? propose methodology for the creation and the description of explicit links in multimedia collections, study the impact of linked media, revisiting existing tasks and technology and developing new applications

International and industrial relations

LinkMedia relies on a number of key collaborations with research institutions around the world: National Institute for Informatics (Japan), Univ. van Amsterdam (The Netherlands), Berkeley (USA), Reykyavic University (Iceland), PUC Minas (Brazil), to name the main ones. Via national and international projects, the team has developped a network of industrial collaborations with leading companies (e.g., Orange, Technicolor), institutions (INA) as well as with SMEs (Exalead, LTU) in the multimedia business. Two companies are spinoff of team members, namely Videntifier Technology and Lamark.

Keywords: Multimedia Indexing Knowledge discovery Machine learning Big data Computer vision Natural language processing Speech and audio