Special Sessions

Four Special Session proposals have been accepted for CBMI 2011:

The submissions to these sessions is open (that is, the authors should decide if the topic of their paper is more appropriate for a Special Session or for a regular track one) and will go through the same review procedure as regular papers. If fewer than the needed number of papers are selected, the Special Session will be cancelled, and the accepted papers will be scheduled in regular sessions.

All accepted and registered papers will be published in the workshop proceedings which will be indexed and distributed by the IEEExplore.

Special Session on Indexing and Retrieval of Multimedia Cultural Heritage Objects (ASSETS)


Cultural heritage (CH) content in multiple media is everywhere, in more traditional environments such as libraries, museums, galleries and audiovisual archives, but also in popular magazines and newspapers. CH objects on the web are no longer isolated objects, but richly connected entities, situated in context, and associated with information from a broad spectrum of sources, representing both authoritative and highly personal views. For both social and economic reasons, there is a growing interest in the development of tools and systems to manage and retrieve cultural heritage content. Europeana, a platform for search over a vast collection of European digital libraries with digitised paintings, books, films and archives is an important example of the effort dedicated to this area by the European Commission.

Digital representations of cultural heritage objects are heterogeneous in nature; they are available in the form of images, videos, audios, 3D models, texts. Making on-line cultural heritage more accessible involves defining new smart techniques and paradigms for indexing and retrieving these objects. Information retrieval techniques applied to CH material must take into account the specificity of the digital representation, the characteristics of the content being indexed and retrieved, and the most suitable search paradigms for the user.

ASSETS is a project funded by the European Commission which aims at developing advanced search services for on-line cultural heritage material to be used in Europeana. The aim of this special session is to bring together scientist and practitioners working in the fields of content-based retrieval and cultural heritage, inside and outside of the ASSETS consortium, to present, propose, and discuss issues and solutions for indexing and retrieving multimedia cultural heritage objects.

Scope and topics

Papers presenting solutions, proposing new indexing and retrieval paradigms, analyzing existing approaches are solicited, including, but not limited to, the following topics:

  • Image indexing and retrieval technique
  • Video indexing and retrieval techniques
  • Video summarization
  • Audio indexing and retrieval techniques
  • Audio summarization
  • 3D models indexing and retrieval
  • Text indexing and retrieval
  • Metadata indexing and retrieval
  • Automated metadata generation
  • Automated annotation
  • Multimedia content extraction
  • Interfaces for multimedia retrieval
  • Presentation and visualization tools
  • Query log analysis
  • Post Query processing
  • User interaction and relevance feedback
  • Large scale indexes for content based retrieval


  • Giuseppe Amato, ISTI - CNR, Italy
  • Víctor Valdés, VPULab - UAM, Spain
To top of the page

Special Session on Collaborative Learning for Interactive and Long Term Retrieval in Multimedia Databases (CLILTR)


Machine learning techniques are now widely used in Content-Based Multimedia Indexing (CBMI) systems to bridge the semantic gap. Most of these methods are designed around a single user, who is invited to build a query made of image annotations. Depending on the difficulty of the searched concept, the user gives more and more annotations, until he is satisfied by the results. In such a scenario, better retrieval systems are those which return better results with less annotations.

Then, once such systems are deployed, many users perform their searches independently. However, it is common that several users are looking for similar visual concepts. That means that a lot of knowledge can be shared among users. In other words, if previous users already found several concepts, a smart retrieval system should take advantage of this previous knowledge in order to speed up the next retrieval sessions.

This problematic is poorly addressed by the content based multimedia retrieval community. The aim of this special session is to first present the latest methods in this scope, but also to invite new researchers to address these learning problems.

Scope and topics

The scope of this session is to cover innovative indexing systems which break the single user paradigm. Such systems are facing many learning problems. In that case, the main question is how to efficiently combine these learning tasks to improve the results. A first example would be the collaborative learning of a single concept and a second one the long-term learning of unknown concepts.

The special session seeks for papers describing original work in the following areas:

  • multi-users retrieval systems
  • collaborative learning of a single concept
  • long-term learning of unknown concepts
  • fusion of retrieval sessions
  • concepts discovery
  • active learning in collaborative learning context


  • Philippe Henri Gosselin, ETIS Lab - ENSEA, France
  • David Picard, ETIS Lab - ENSEA, France
To top of the page

Special Session on Event-based multimedia indexing (EBMI)


With the rapid progress of hardware technology and the popularity of related multimedia devices, the quantity of multimedia data has surged into an unprecedented level. These data reside on local personal computers or global large scale repositories, and large amounts of them are daily consumed within the framework of networked media. For manipulating this information, sophisticated algorithms are needed for supporting the automatic indexing of multimedia. However, this is to some extent still beyond the capabilities of the current state-of-the-art in multimedia management. An important reason for this deficiency is that machines typically index multimedia content on the basis of low-level features, or at best on the basis of concepts corresponding to broad media classes or tangible objects. In contrast to this, there has recently been strong argumentation in favor of the conception that humans remember real life using past experience structured in events. For this reason, the development of methods that approximate the human perception of events and support the automatic or semi-automatic event-based indexing of multimedia content recently emerged as a very promising research direction.

Scope and topics

The special session welcomes high-quality papers describing original research in all areas related to the event-based indexing of multimedia, including:

  • Detection of events in multimedia content
  • Indexing structures and multimedia event models
  • Interactive tools for event-based media organization
  • Event-based search and retrieval
  • Event-based applications


  • Vasileios Mezaris, CERTH - ITI, Greece
  • Nicu Sebe, University of Trento, Italy
  • Francesco G. B. De Natale, University of Trento, Italy
To top of the page

Special Session Social Media Retrieval and Filtering (SMRF)


Nowadays, an ever increasing amount of multimedia content is available on content sharing websites such as Flickr, Youtube or Facebook. This content consists of text and multimedia data, enriched with additional metadata (annotations, comments, tags, etc.). In addition, semantic relationships between content and content sharers can be inferred by considering the structure of the social content sharing websites. However, this social factor has hardly been studied in the information retrieval and recommendation domain. This special session aims at state-of-the-art research approaches that address the underlying social structure for retrieval and filtering of multimedia content.

Scope and topics

We are interested in works in multimedia retrieval and filtering on the Social Web, including, but not limited to the following research topics:

  • Analysis, indexing, search and retrieval of social media content
  • Scalable indexing and structure of social media content
  • Semantic-based retrieval and filtering of social media content
  • Learning, personalization, content adaptation, and relevance fedback in social media systems
  • Personal multimedia content management in social media systems
  • Social and content-based multimedia recommendation
  • Summarization, organization, browsing, and exploration of social media content
  • User interfaces for management, presentation and visualization of social media content
  • Social media data indexing, search and retrieval in mobile environments
  • Context-awarenes in social media systems


  • Frank Hopfgartner, ICSI, Berkeley, USA
  • David Vallet, IRG - UAM, Spain
  • Iván Cantador, IRG - UAM, Spain
To top of the page

Universidad Autónoma de Madrid, Escuela Politécnica Superior
Video Processing and Understanding Lab - Information Retrieval Group