Special Sessions


Within CBMI 2013 we plan the following special and demo sessions:


Special Session 1: Visual Search in Broadcast and Video Surveillance Data

Due to the rapidly growing amount of digital multimedia content there is a need for quickly pinpointing relevant data within the “sea” of irrelevant. Manual search or browsing in such large amounts of data is typically not feasible, since it is extremely time-consuming, exhausting, and most likely unsuccessful given that typically relevant data represents only a small fraction of the entire dataset. This special session of the CBMI'13 invites unpublished, original research relating to search and retrieval within broadcast and surveillance video data. More specifically, the topics of interest include (but are not limited to):

- Person re-identification and search,

- Analysis and retrieval of specific image patterns (text, logo, objects),

- Novel feature representation and semantic indexing

within the context of broadcast and surveillance data analysis.


Session Chair’s bio: Csaba Beleznai (AIT Austrian Institute of Technology GmbH)

Csaba Beleznai received his M.S. degree from the Technical University of Ilmenau (Germany) in electrical engineering in 1994. He received his Ph.D. degree in physics from the Claude Bernard University, Lyon (France) in 1999. C. Beleznai joined the research center "Advanced Computer Vision" in 2000 and coordinated research activities in the area of "Surveillance and tracking". From 2007 to 2008 he served as Vice-President of the Austrian Association for Pattern Recognition. Currently he is Senior Scientist at the Austrian Institute of Technology (AIT), where he scientifically coordinates the applied research center Embedded Computer Vision (ECV), which is a task-oriented research project aiming at the joint development of computer vision algorithms and embedded hardware concepts for applications in demanding industrial context.



Special Session 2: Web-based (semantic) indexing

This special session focuses on the following topics:

- Web multimedia content acquisition and mining.
- Automated multimedia description from Web resources and crowdsourcing.
- Studies of multimedia information-seeking behavior.
- Community-based multimedia content management.
- High performance multimedia indexing algorithms, scalability of multimedia information retrieval systems.


Session Chair’s bio: András Benczúr (Data Mining and Search Group at MTA SZTAKI)

András Benczúr is the head of Informatics Laboratory of 20 doctoral students, post-docs and developers hosting in addition a Theory of Computing and a Natural Language Technology groups. Andras received his Ph.D. at the Massachusetts Institute of Technology in 1997, since then his interest turned to Information Retrieval and Web Search. He was representing SZTAKI as principal investigator in several EU and national R&D projects. His research on spam filtering and low space approximations for very large scale Web analysis was awarded by a Yahoo Faculty Research Grant in 2006. Andras is SZTAKI site coordinator in the Hungarian Future Internet Platform and in the Hungarian node of the FET Flagship FuturICT.


Demo session

CBMI 2013 is looking for submissions of original, state of the art (and beyond) Demos (4 page demo paper + demonstration) to be included in the CBMI 2013 Demo Session. The demos may be of content-based multimedia
indexing systems, as well as of other systems related to the topics of the workshop.

Demo proposals should include:
1. Max. 1 page cover letter describing the to-be demonstrated method's/system's topic, the requirements to set up the demonstration (what you will bring, and what would be required to be provided by the local organizers), and a description of the demonstration itself. URLs of videos showcasing the demonstration would be welcome (but optional).

2. A 4 page demo paper, as described in the Instructions section.

Upon acceptance the demo paper will be included in the conference and proceedings.

Demo papers and cover letter should be submitted through the CBMI 2013 EasyChair site, for dates see the Conference Call section of the CBMI 2013 website.

Session Chair’s bio: Levente Kovács (Distributed Events Analysis Research Laboratory  at MTA SZTAKI)

Levente Kovacs received his MSc in IT (2002) and PhD in image processing and graphics (2007) from the University of Pannonia, Hungary. His main research areas are image/video feature extraction and indexing, annotation, event detection, non-photorealistic rendering, and video restoration. Currently working as senior research fellow at the Distributed Events Analysis Research Laboratory of the Computer and Automation Research Institute of the Hungarian Academy of Sciences (MTA SZTAKI), Budapest, Hungary.


Special Session 3: Challenges in Multimedia indexing and Retrieval

Evaluation campaigns (or challenges or benchmarks) are very important for comparing approaches and measuring progress over time. Many such campaigns are now available in multimedia indexing and retrieval, most of them being issued yearly. Beyond the evaluation aspect, they are also used for directing researches in the field in promising directions. This special session of the CBMI'13 invites papers related to such evaluation campaigns.

More specifically, the topics of interest include (but are not limited to):
- the organization of challenges or tasks in multimedia indexing and retrieval;
- the analysis or review of results of challenges or tasks in multimedia indexing and retrieval;
- constitution of relevant datasets with associated annotations;
- evaluation protocols or metrics for the evaluation of multimedia indexing and retrieval systems.

Session Chair’s bio: Dr. Georges Quénot (CNRS, French National Centre for Scientific Research)

Dr. Georges Quénot is senior researcher at CNRS (French National Centre for Scientific Research). He has  a PhD in Computer Science (1988) and a Habilitation à Diriger des Recherches in Computer Science (1998) from the University of Orsay. He is currently leading the Multimedia Information Indexing and Retrieval group (MRIM) of the Laboratoire  d'Informatique de Grenoble (LIG) where he is also responsible for their activities on video indexing and retrieval. His current research activity is about semantic indexing of image and video documents using supervised learning, networks of classifiers and multimodal fusion. He is responsible of the semantic indexing task at TRECVID since 2010.

Monday the 4th. .