MediaEval is a multimedia benchmark evaluation that offers tasks promoting research and innovation in areas related to human and social aspects of multimedia. MediaEval 2012 focuses on aspects of multimedia including and going beyond visual content, such as, speech, language, audio and social factors. Participants carry out one or more of the tasks offered and submit runs to be evaluated. They then write up their results and present them at the MediaEval 2012 workshop.
For each task, participants receive a task definition, task data and accompanying resources (dependent on task) such as shot boundaries, keyframes, visual features, speech transcripts and social metadata. In order to encourage participants to develop techniques that push forward the state-of-the-art, a "required reading" list of papers will be provided for each task. Participation is open to all interested research groups. Please sign up via http://www.multimediaeval.org (regular sign up will remain open until 31 May).
The following tasks are available to participants at MediaEval 2012:
Placing Task This task involves automatically assigning geo-coordinates to Flickr videos using one or more of: Flickr metadata, visual content, audio content, social information (Data: Creative Commons Flickr data, predominantly English language, extended from the 2011 data set.)
Social Event Detection Task This task requires participants to discover events and detect media items that are related to either a specific social event or an event-class of interest. By social events we mean that the events are planned by people, attended by people and that the social media are captured by people. (Data: URLs of images and videos available on Flickr and other internet archives together with metadata).
Spoken Web Search Task This task involves searching FOR audio content WITHIN audio content USING an audio content query. It is particularly interesting for speech researchers in the area of spoken term detection. (Data: Audio from four different Indian languages and four South African languages. Each of the ca. 2000 data item is an 8 KHz audio file 4-30 secs in length.)
Tagging Task Given a set of tags and a video collection, participants are required to automatically assign the tags to each video based on a combination of modalities, i.e., speech, metadata, audio and visual. (Data: Creative Commons internet video, nearly exclusively English, extended from the 2011 collection.)
Affect Task: Violent Scenes Detection This task requires participants to deploy multimodal features to automatically detect portions of movies containing violent material. Any features automatically extracted from the video, including the subtitles, can be used by participants. (Data: A set of ca. 18 Hollywood movies that must be purchased by the participants.)
Visual Privacy Task For this task, participants propose methods whereby human faces occurring in digital imagery can be obscured so as to render them unrecognizable. An optimal balance should be struck between obscuring identity and maintaining the quality of the viewing experience from the user perspective. (Data: about 100 high resolution video files of ca 1m30s each and containing one or more persons in an indoor environment.)
Brave New Tasks This year, MediaEval will also run three new tasks in the areas of social media, spoken content search and hyperlinking, and music tagging. These tasks are "by invitation only" and are not included in the general registration form. In order to receive an invitation, please contact the task organizers.
MediaEval 2012 Timeline (dates vary slightly from task to task, see the individual task pages for the individual deadlines: http://www.multimediaeval.org/mediaeval2012)
31 May: Last day for regular sign up 1 June: Latest day for development data release 1 July: Latest day for test data release ca. 10 September: Run submission deadline 28 September: Working notes papers due 4-5 October: MediaEval 2012 Workshop, Pisa, Italy* *The workshop is timed so that it is possible to attend the 12th European Conference on Computer Vision ECCV 2012 (http://eccv2012.unifi.it/), held 7-13 October in Firenze, Italy, in the same trip.
MediaEval 2012 Coordination Martha Larson, Delft University of Technology Gareth Jones, Dublin City University
Contact For questions or additional information please contact Martha Larson m.a.larson at tudelft.nl or visit visit http://www.multimediaeval.org
MediaEval 2012 Organization Committee:
Robin Aly, University of Twente, Netherlands Xavier Anguera, Telefonica, Spain Atta Badii, University of Reading, UK Etienne Barnard, CSIR, South Africa Claire-Helene Demarty, Technicolor, France Maria Eskevich, Dublin City University, Ireland Gerald Friedland, ICSI, USA Isabelle Ferrané, University of Toulouse, France Guillaume Gravier, IRISA, France Claudia Hauff, TU Delft, Netherlands Gareth Jones, Dublin City University, Ireland Pascal Kelm, Technical University of Berlin, Germany Christoph Kofler, Delft University of Technology, Netherlands Chattun Lallah, University of Reading, UK Martha Larson, TU Delft, Netherlands Cynthia Liem, TU Delft, Netherlands Florian Metze, CMU, USA Vasileios Mezaris, ITI Certh, Greece Roeland Ordelman, University of Twente and Netherlands Institute for Sound and Vision, Netherlands Nicola Orio, Università degli Studi di Padova, Italy Geoffroy Peeters, Institut de Recherche et Coordination Acoustique/Musique Paris, France Cedric Penet, Technicolor, France Tomas Piatrik, Queen Mary University of London, UK Adam Rae, Yahoo! Research, Spain Nitendra Rajput, IBM Research, India Markus Schedl, Johannes Kepler Universität Linz, Austria Sebastian Schmiedeke, Technical University of Berlin, Germany Mohammad Soleymani, University of Geneva, Switzerland Robin Sommer, ICSI/LBNL, USA Raphael Troncy, Eurecom, France
A large number of projects make a contribution to MediaEval organization, including (alphabetically): AXES (http://www.axes-project.eu), Chorus+ (http://www.ist-chorus.org), CUbRIK (http://www.cubrikproject.eu/), Glocal (http://www.glocal-project.eu), IISSCoS (http://www.cdvp.dcu.ie/IISSCoS/), LinkedTV (http://www.linkedtv.eu/), Promise (http://www.promise-noe.eu/), Quaero (http://www.quaero.org), Sealinc Media (http://www.commit-nl.nl/), VideoSense (http://www.videosense.eu/) and SocialSensor (http://www.socialsensor.org/).