The Lab has extensive experience and expertise in semantic multimedia analysis, indexing and retrieval, social media and big data analytics, knowledge structures, reasoning and personalization for multimedia applications, eHealth and environmental applications

Best Poster Award @ ACM ICMR 2017

The paper “Concept Language Models and Event-based Concept Number Selection for Zero-example Event Detection”, by D. Galanopoulos, F. Markatopoulou, V. Mezaris, and I. Patras, received the Best Poster Award at the ACM International Conference on Multimedia Retrieval (ICMR 2017) that took place in Bucharest, Romania.

MKLab will present their work on Learning to Detect Misleading Content on Twitter and YouTube to ICMR 2017

MKLab have been working for some time now on the problems of fake news and of misinformation in social media. Our work has recently led to some important achievements: we developed a very effective machine learning-based algorithm for automatically deciding whether a tweet is credible, i.e. whether it accurately reflects the event to which it refers. Training this algorithm became possible by creating a large corpus of past tweets that have been annotated as “fake” or “real”.

MKLab Published its First MOOC on Udemy

MKLab proudly announces the release of its "Dynamics of Knowledge Organisation" MOOC on Udemy. Featuring more than 2 hours of recorded video content, the course studies the involvement and application of semantics in Digital Preservation and has been produced within the context of the PERICLES FP7 project. The MOOC is taught by Prof. Sándor Darányi (School of Library and Information Science, University of Borås) and Dr. Efstratios Kontopoulos (MKLab, CERTH/ITI), and is supported by Mr. Panagiotis Mitzias (MKLab, CERTH/ITI).

Release of concept detection scores for IACC.3

We provide a complete set of concept detection scores, using our state of the art concept detectors, for the videos of the IACC.3 dataset that is used in the TRECVID AVS Task from 2016 and on (600 hr of internet archive videos). Concept detection scores for 1345 concepts (1000 ImageNet concepts and 345 TRECVID SIN concepts) have been generated using two different automatic video annotation methods. The dataset and the related publications can be found at: https://www.

MKLab successfully showcased the results of HOMER Project to Law Enforcement Agents from South East Europe

The HOMER South East European RoadShow Event was held in Thessaloniki, Greece on December 15, 20016, co-organized by MKLab (ITI-CERTH), South-East European Research Centre and the Bulgarian Defense Institute. During the event the achievements of the HOMER consortium were presented to law enforcement and security agencies from South East Europe focusing on demonstrating the Homemade Explosive (HME) Discovery Tools and the HME Knowledge Management Platform (KMP). The event welcomed about 30 law enforcement and security agents representing several organizations across South East Europe, including Interpol, the Hellenic Police, the Hellenic Ministry of Defense, the Bulgarian Army, the Bulgarian Police, the Cyprus Police, the Romanian Armed Forces, the FYROM Ministry of Interior, the Bulgarian Ministry of Defense, the Serbian Ministry of Defense, the Albanian Police, the Bosnian Police and the UK Embassy in Greece.

Special Track - MultiModal Interfaces for Natural Human Computer Interaction: Theory and Applications

A special track on MultiModal Interfaces for Natural Human Computer Interaction: Theory and Applications is organized in conjuction with the 30th IEEE International Symposium on Computer-Based Medical Systems - IEEE CBMS 2017, which will take place in Thessaloniki, Greece from Thursday 22nd to Saturday 24th of June 2017. Important dates: Paper submission due (general and special tracks) January 30, 2017 Notification of acceptance for papers March 20, 2017 Final camera-ready paper due April 13, 2017 Early registration deadline April 23, 2017 IEEE CBMS 2017 conference days June 22-24, 2017

MKLab Participating in PERICLES PhD Course

MKLab enthusiastically participates in the PhD course "Dynamics of Knowledge Organization" that is funded by the PERICLES FP7 project and is organized by the School of Library and Information Science, University of Borås.   The 2-month course started on 6-Dec and involves 8 students from around Europe, who will undergo intensive lectures, seminars and labs, where they will also have the opportunity to play around with PERICLES software tools.    The course focuses on the nature and role of knowledge organization in a rapidly changing digital world.

MKLab Showcases the Outcomes from the REVEAL Project

The REVEAL project is nearing its end (official end date is 31 December 2016), and our team has a number of interesting results to share with the community. The results are of interest for a number of applications related to news mining, social media retrieval and verification. In particular, we make available: a comprehensive toolbox for image forensics along with a web-based user interface; a method and web user-interface for verifying tweets; a web service for automatically detecting images with disturbing content; a method for predicting the future popularity of reddit posts and YouTube videos; a method for classifying Twitter users based on their interactions with other users; a method for estimating the geographic location referred to by a piece of text; a method for automatically creating summaries from large collections of social media posts.

MKLab and INAB joint work on individualized modelling in CLL (hematologic cancer) accepted as a project in European Research Initiative on CLL (ERIC)

ERIC, the European Reasearch Initiative on Chronic Lymphocytic Leukemia (CLL), has accepted our joint work with the Institute of Applied Bioscience (INAB) on individualized modelling of CLL patients, presented at ASH 2015 [1]. Our objective in this work was to compare the performance of standard prognostic models in CLL (Cox regression models), novel machine learning models and clinicians' prognosis in predicting the Time-To-First-Treatment (TTFT) for each patient. ERIC accepted our work as an ongoing project with the objective to promote it to trained hematologists in order to obtain a large set of educated predictions with respect to TTFT.

MKLab participation to the TRECVID 2016 AVS task

MKLab successfully participated to the Ad-hoc Video Search (AVS) of TRECVID 2016. The AVS task attempts to model the end-user video search use-case, where the user is looking for segments of video containing persons, objects, activities, locations, etc. and combinations of the former. To express this information need, the user provides to the retrieval system a short query in natural language, e.g., “Find shots of a person playing guitar outdoors”. The experiments this year were performed on a set of Internet Archive videos totalling about 600 hours of video duration, and using 30 different queries.

BDV associate member

Improve My City Mobile

imc logo

Improve My City Mobile, allows citizens to report local problems and suggest solutions for improving their neighbourhood. Learn more...

Motorola collaborations

Motorola Logo

The Multimedia Group has been collaborating with Motorola in several Motorola funded R&D projects.

GPU-LIBSVM

NVIDIA links Multimedia Group for the GPU-LIBSVM implementation

HR Excellence in Research

HR Excellence in Research