3rd International Workshop on Linked Media

*** TWITTER HASHTAGS #lime2015 AND #linkedmedia ***

This full day workshop was co-located with the WWW 2015 conference held in Florence, Italy.

It took place on the 18th May 2015.

The workshop proceedings are available here.

Workshop Program

14:00: Welcome by the chairs
14:05: Keynote

Harald Sack, HPI Potsdam

“From Script Idea to TV Rerun – The Idea of Linked Production Data in the Media Value Chain”

Within the process of the production of a film or tv program a significant amount of metadata is created and – most times – lost again. As a consequence most of this valuable information has to be costly recreated in subsequent steps of media production, distribution, and archival. On the other hand, there is no commonly used metadata exchange format throughout all steps of the media value chain. Furthermore, technical systems and software applications used in the media production process often have proprietary interfaces for data exchange. In the course of the D-Werft project funded by the German government, metadata exchange through all steps of the media value chain is to be fostered by the application of Linked Data principles. Starting with the idea for a script, metadata from existing systems and applications will be mapped to ontologies to be reused in subsequent production steps. Also for distribution and archival, metadata collected during the production process is a valuable asset to be reused for semantic and exploratory search as well as for intelligent movie recommendation and customized advertising.

Harald Sack is Senior Researcher at the Hasso Plattner-Institute for IT-Systems Engineering (HPI) at the University of Potsdam and head of the research group ‘Semantic Technologies and Multimedia Retrieval’. His current areas of research include semantic technologies, knowledge discovery as well as multimedia analysis & retrieval. He is co-initiator of the D-Werft project, funded by the German ministry of education and research, focussed on semantic data integration in the media value chain (2014-2017). He is co-founder of yovisto, which successfully markets state-of-the-art semantic video analysis & search technologies.

14:45: “Defining and Evaluating Video Hyperlinking for Navigating Multimedia Archives”

Roeland Ordelman, Maria Eskevich, Robin Aly, Benoit Huet and Gareth Jones

15:15: Demo “The TIB|AV Portal as a future Linked Media Ecosystem”

Paloma Marín Arraiza and Sven Strobel

Demo 1
Demo 2
Demo 3



16:00: “Enabling access to Linked Media with SPARQL-MM”

Thomas Kurz, Kai Schlegel and Harald Kosch

16:30: Demos “MICO — Towards Contextual Media Analysis”

Sergio Fernández, Sebastian Schaffert and Thomas Kurz


and “Automating Annotation of Media with Linked Data Workflows”

Thomas Wilmering, Sean Bechhofer, Simon Dixon, György Fazekas and Kevin Page


17:00: Group Activity: bringing Linked Media to life!


18:00: Wrap up and close


  • Submission deadline: Feb 7, 2015 (23:59 Hawaii Standard Time)
  • Paper acceptance notification: Feb 22, 2015
  • Paper final copy hard deadline: Mar 8, 2015

Goals of the workshop

If the future Web will be able to fully use the scale and quality of online media, a Web scale layer of structured media annotation is needed, which we call Linked Media. This 3rd international workshop on Linked Media (LiME 2015), building on two successful events, aims at promoting the principles of Linked Media on the Web by gathering semantic media researchers, media owners, and media service providers to exchange current research and development work on online media description creation, publication, and processing. Specifically, we aim to promote the core principles of Linked Media and explore their value through demonstrations: the online publication of structured media descriptions, making media more easily shared, queried and re-used. This will offer a wide range of possibilities for various stakeholders in the creative industries.

Of specific relevance to the WWW community is the fact LiME offers a platform to demonstrate application of the work in and inform ongoing working groups in this area: Web&TV Interest Group and Presentation API working group (dealing with multimedia in HTML5) and the W3C Annotations working group (focusing on the Open Annotation Model). There’s great value of encouraging cross-fertilization between these working groups and with the broader WWW community.

Topics and themes

Today’s Web is a rich media Web – non-textual content is often now the first destination of online agents rather than HTML/textual resources. As a result, access to structured annotation of the online media is increasingly important for new Web applications capable of media search, retrieval, adaptation and presentation. Yet, the online media annotation space is still limited, fragmented and lacking in consensus for building Web tools and interfaces to support it. The W3C Ontology for Media Resources provides mappings between 18 different multimedia metadata schema or standards and established a first step towards a common schema model which now requires championing in the research and industry communities. The least common denominator approach followed by the W3C group has lead to a small and useful vocabulary that fails to support more advanced use cases that require to describe the multimedia content at a fragment level and beyond simple tagging. We see opportunity in the uptake of the Open Annotation Model within a W3C Annotations WG and want to promote its usage for media description combined with Linked Data.

If the future Web will be able to fully use the scale and quality of online media, a Web scale layer of structured media annotation is needed, which we call Linked Media, which is inspired by the Linked Data movement for making structured descriptions of resources more available online. Mobile and tablet devices, as well as connected TV introduce novel application domains that benefit from broad understanding and acceptance of Linked Media standards. LiME 2015 aims at promoting the principles of Linked Media on the Web by gathering media owning stakeholders and semantic media researchers to exchange current research and development work on online media description creation, publication, and processing.

Important aspects to discuss revolve around (1) emerging approaches to online media description, (2) extracting such descriptions and linking them to external resources, and (3) showcasing practical use cases in this domain, also covering interaction aspects for single and group users.

The workshop topics include, but are not limited to:

1. Approaches to online media descriptions & annotation

  • Aligning the fragmented approaches to online media description, its publication, and processing;
  • Tools and approaches to lower the cost of creating structured descriptions of online media resources;
  • New methods of automatic, real time, metadata extraction of any online media content (including live streams);
  • Ideas how to incorporate Linked Data into media description (and benefit from the additional metadata of the Linked Data cloud);
  • Use of the Open Annotation model for describing online media.

2. Methods for enriching and hyperlinking media

  • Tools and approaches to search and retrieval of online media based on its structured description, scaling to the Web;
  • New methods for automatically assessing the suitability of (non-trusted) content for interweaving (e.g. violence detection, nudity detection), and publishing such assessments;
  • Methods and tools for automatically link media to media at the fragment level, including evaluation on standard corpora such as MediaEval;
  • Addressing issues of trust, quality and rights of online media.

3. Presentation, interaction, evaluation and new business models for Linked Media

  • New Web applications making use of Linked Media, also across different platforms;
  • Evaluations of innovative services with end-users;
  • Approaches to tracking user interaction with media (and exploiting this knowledge to enrich annotations);
  • Emergence of new business opportunities.


Submissions should not exceed 6 pages and are to be formatted according to the ACM SIG proceedings template (http://www.acm.org/sigs/publications/proceedings-templates) and submitted to https://www.easychair.org/conferences/?conf=lime2015. Papers should be submitted in PDF format. We encourage various types of submission:

  • full papers (max 6 pages) for mature work which has been subject to evaluation
  • demo submissions (max 2 pages) for demos, software and platforms which may be able to support a part of the Linked Media ecosystem

The workshop proceedings will be published in the ACM Digital Online Library as part of the companion volume of the WWW proceedings.


Programme Committee

  • Olivier Aubert, University of Nantes
  • Werner Bailer, Joanneum Research
  • Louay Bassbouss, Fraunhofer FOKUS
  • Marco Bertini, Universita’ degli Studi di Firenze
  • Tobias Bürger, PAYBACK
  • Pierre-Antoine Champin, LIRIS
  • Jean-Claude Dufourd, Telecom Paristech
  • Antoine Isaac, Europeana & VU Amsterdam
  • Yunjia Li, University of Southampton
  • Erik Mannens, Ghent University
  • Giuseppe Rizzo, EURECOM
  • Harald Sack, Hasso Plattner Institute Potsdam
  • Thomas Steiner, Google
  • Davy Van Deursen, EVS
  • Ruben Verborgh, Ghent University – iMinds
Leave a Comment

You must be logged in to post a comment.