ACM International Conference on Image and Video Retrieval
 
 
July 8-10 2009, Island of Santorini, Greece
 
 

VideOlympics Showcase

Demo sessions of video retrieval systems are ideal venues to disseminate scientific results. Existing demo sessions, however, fail to engage the audience fully. Real-time evaluation of several video retrieval systems in a single showcase increases impact. Encouraged by the success of previous editions, we will again organize a VideOlympics showcase at the ACM International Conference on Image and Video Retrieval.

The major aim of the VideOlympics is to promote video retrieval research. An additional main goal of the VideOlympics is giving the audience a good perspective on the possibilities and limitations of the current state-of-the-art systems. Where traditional evaluation campaigns like TRECVID focus primarily on the effectiveness of collected retrieval results, the VideOlympics also allows to take into account the influence of interaction mechanisms and the advanced visualizations in the interface. Specifically, we aim for a showcase that goes beyond the regular demo session: it should be fun to do for the participants and fun to watch for the conference audience. For all these reasons, the VideOlympics should only have winners. Similar to previous years, a number of TRECVID participants will simultaneously do an interactive search task during the VideOlympics showcase event.

New in 2009

For the first time, we will include in the 2009 edition of the VideOlympics a round with novice users, in addition to the round with expert users. The novice users will be selected from a group of high-school teenagers from the island of Santorini, for whom it can be assumed that they have a decent English language level. Moreover it is allowed to provide each novice user with a short training session with your video search engine (amount to be defined).

Setup

We will be working with TRECVID 2008 test data (100 hrs of broadcast TV news, documentaries, and educational programming) from the Netherlands Institute for Sound and Vision. Text-only search topics will be revealed during the showcase, similar in spirit to those used in several editions of the TRECVID interactive search tasks, e.g.:

  • Find shots of traffic signs
  • Fins shots of a meeting with a large table and people visible
  • Find shots of a daytime demonstration or protest with at least part of one building visible

In contrast to TRECVID's interactive search task, where results are submitted at the end of the search session, results should be submitted immediately after they are found (see our protocol). Thus encouraging quick retrieval of relevant results as well as unique results not found by others. Note that the main aim of the VideOlympics is giving the audience a good perspective on the possibilities and limitations of the current state-of-the-art systems. Results are not meant for publication, but for winning the Golden Retriever awards. Golden Retrievers will be awarded for:

  • Best performer
  • Most impressive interface
  • The public’s favorite
  • The one the public could use
  • ...
In summary, the VideOlympics will only have winners.

We will provide a score board to allow for evaluation on the spot, which will be based on the results and on votes by the audience.

Constraints

  • Only one system per team can be connected to the scoreboard
  • Your interactive systems should work on at least one laptop, with all relevant TRECVID 2008 test data on board.
  • Your interactive system can not use on-line information extraction from the Internet.
  • We provide the protocol and software for a simple submit system of results.

Participation

It is no longer possible to submit proposals for participation in the 2009 VideOlympics.

Please include the following:
  • A 1-page description of your system, possibly including a screen shot, formatted in PDF using the camera-ready templates available here.
  • Indicate whether you will participate in the novice and/or expert round.
  • 3 possible text-only search topics of varying complexity (NOT similar to existing TRECVID 2007/2008 topics).
All system descriptions will be peer-reviewed by at least 2 referees. Criteria for selection are novelty, uniqueness, and potential for audience involvement. If your system is accepted, you are required to provide ground truth annotation for 1 of your submitted search topics (the topic will be selected by the organization).

Organizers

Cees Snoek, University of Amsterdam, The Netherlands
Marcel Worring, University of Amsterdam, The Netherlands
Rong Yan, IBM Research, USA
Alex Hauptmann, Carnegie Mellon University, USA

Co-Organizers

Ork de Rooij, University of Amsterdam, The Netherlands
Koen van de Sande, University of Amsterdam, The Netherlands
Stefanos Vrochidis, Informatics and Telematics Institute, Greece