MKLab

The paper  "The Dynamics of Content Popularity in Social Media" by Symeon Papadopoulos, Athena Vakali and Yiannis Kompatsiaris received the best 2010 paper award by the International Journal of Data Warehousing and Mining (IJDWM) and will be nominated for the Annual Excellence in Research Journal Award in Database Technologies.  Article pdf. Journal website.

A GPU-accelerated implementation of LIBSVM was released. This is a modification of the original LIBSVM that exploits the CUDA framework to significantly reduce processing time while producing identical results. GPU-accelerated LIBSVM gives a performance gain depending on the size of the input data set; this gain increases significantly when considering large data sets. More details and the software can be downloaded from http://mklab.iti.gr/project/GPU-LIBSVM

A book chapter on Collective Intelligence titled "Emerging, Collective Intelligence for personal, organisational and social use" has been accepted for publication in Nik Bessis and Fatos Xhafa (eds.) Next Generation Data Technologies for Collective Computational Intelligence, “Studies in Computational Intelligence” book series, Springer (2011).  The chapter was written by Sotiris Diplaris, Andreas Sonnenbichler, Tomasz Kaczanowski, Phivos Mylonas, Ansgar Scherp, Maciej Janik, Symeon Papadopoulos, Michael Ovelgönne, and Yiannis Kompatsiaris. It is a joint dissemination result of the WeKnowIt project, presenting Collective Intelligence research work, applications and results from the WeKnowIt perspective.

PatMedia was presented and demonstrated to the Director of the sector of the European patent office on Image Processing and Computer Games, Mr Benno Penzkofer. "This tool could speed up the searches of my patent examinors", were the words of Mr. Penzkofer after the presentation. More detailed information, on line demonstration and more testimonials related to PatMedia are available here..

The ClustTour idea was submitted to the OpenFund -a Greek venture capital aiming at financing and supporting promising high-tech start ups- and received positive feedback with respect to its technical aspects. In addition, it received valuable feedback regarding the potential for exploiting the underlying technologies. Reviewers pointed the utility of the service for travellers, as well as the technical soundness of the idea and the development team. In addition, they recommended the redefinition of the business model such that the prospective business can be sustainable from well-specified (and plausible) revenues.

Our paper "Cluster-based Landmark and Event Detection on Tagged Photo Collections" was accepted for publication in the IEEE MultiMedia magazine. The paper describes the technologies underlying the ClustTour web application that was developed in the context of the WeKnowIt project. Citation: S. Papadopoulos, C. Zigkolis, Y. Kompatsiaris, and A. Vakali. "Cluster-based Landmark and Event Detection on Tagged Photo Collections". In IEEE Multimedia Magazine (accepted for publication, October 2010)

CBMI 2011 is seeking original, high quality proposals for Special Sessions, to be included in the technical program along with the regular track. The call is available on line.  

WeKnowIt was presented in the ICT event 2010 networking session on Web of Events on October 27th, 2010, in Brussels. Yiannis Kompatsiaris gave a presentation entitled WeKnowIt: Event Detection Processing and Representation.

CERTH participated to the Semantic Indexing (SIN) Task of TRECVID 2010 with its video concept detectors. CERTH’s concept detectors produced the best results, among the submissions made by all institutions participating to TRECVID (101 runs from 30 different institutions), for two of the concepts evaluated this year ("walking", "sitting down"). The latest technical details on the approach that was developed and used for building these detectors are reported in the following publication: V.