Dr Tomas Piatrik

Contact Details

Research Group: Multimedia and Vision Research Group

Research Topic: Image Clustering/Classification using Ant Colony Optimization Systems

Research Scope

In the past decade, there has been a rapid growth in the consumption of digital media, specifically, video and images. Before the emergence of content-based retrieval, media was manually annotated with text, allowing the media to be accessed by text-based searching. However, manual annotation is an expensive and labor intensive procedure. Hence automatic techniques for video and image annotation, search and retrieval are critically needed. A fundamental step towards content based image and video annotation and retrieval is classification or clustering. Although there have been several proposals to cope with the underlying classification task, little research has been done on the use of bio-inspired algorithms for the classification of visual data.

Some recent studies have pointed that, the self-organization of neurons into brain-like structures, and the self-organization of ants into a swarm are similar in many respects. Ants present a very good natural metaphor to evolutionary computation. With their small size and small number of neurons, they are not capable of dealing with complex tasks individually. The ant colony in the other hand is many times seen as an "intelligent entity" for its great level of self-organization and the complexity of tasks it performs. Their colony system inspires many researchers in the field of Computer Science to develop new solutions for optimization and artificial intelligence problems.

Research interests: Machine Learning, Biologically Inspired Systems, Semantic Multimedia.

Participation in projects:

Biography

Tomas Piatrik obtained his PhD in the Multimedia and Vision (MMV) Research Group in 2009. His studies concerned image clustering, video summarisation and scene detection based on ant-inspired optimization techniques. He is currently working as a postdoctoral researcher in the MMV group and his research interests include multimedia analysis and retrieval, knowledge extraction, machine learning, biologically inspired computing and future media internet. He has published over 30 technical papers and reports in various international conferences and book chapters. He has actively participated in several EU funded research projects including K-Space, MESH, PetaMedia, 3DLife and support actions SALA+ and Eternals. His current role in MMV is to support research collaboration and standardisation activities in networked media scientific fields, especially in Future Media Internet, as a part of the support/coordination action NextMEDIA. Furthermore, he is coordinating MMV activities in the VideoSense project on privacy-respecting Video Analytics in Security.

Publications

Book chapters

• S. Essid, M. Campedel, G. Richard, T. Piatrik, R. Benmokhtar, and B. Huet, "K-Space Book: Machine Learning Techniques for Multimedia Analysis," 2008,

Conferences

• G. Passino, T. Piatrik, I. Patras, and E. Izquierdo, "A Multimedia Content Semantics Extraction Framework for Enhanced Social Interaction," in Proc. EuroITV 2009 workshop on Enhancing Social Communication and Belonging by Integrating TV Narrativity and Game-Play, 2009,
• T. Piatrik and E. Izquierdo, "Hierarchical Summarisation of Video using Ant-Tree Strategy," in Proc. Proc. of 7th International Workshop on Content-Based Multimedia Indexing, 2009,
• E. Dumont, B. Merialdo, S. Essid, W. Bailer, H. Rehatschek, D. Byrne, H. Bredin, N. O'Connor, G. Jones, A. F. Smeaton, M. Haller, A. Krutz, T. Sikora, and T. Piatrik, "Rushes Video Summarization Using a Collaborative Approach," in Proc. TRECVID BBC Rushes Summarization Workshop at ACM Multimedia, 2008,
• Q. Zhang, K. Chandramouli, T. Piatrik, G. Tolias, U. Damnjanovic, B. Mansencal, A. Saracoglu, N. Aginako, A. A. Alatan, L. A. Alexandre, Y. Avrithis, J. Benois-Pineau, M. Corvaglia, E. Esen, A. Dimou, I. Garcia, A. Hanjalic, R. Jarina, P. Kapsalas, I. Kompatsiaris, L. Makris, V. Mezaris, A. Moumtzidou, P. Mylonas, U. Naci, E. Spyrou, S. Vrochidis, N. Fatemi, F. Guerrini, P. King, and P. Migliorati, "COST292 experimental framework for TRECVID 2008," in Proc. Proc. of TRECVid Workshop, 2008,
• T. Piatrik and E. Izquierdo, "An Application of Ant Colony Optimization to Image Clustering," in Proc. K-Space Jamboree Workshop, 2008,
• E. Izquierdo, K. Chandramouli, M. Grzegorzek, and T. Piatrik, "K-Space Content Management and Retrieval System," in Proc. 14th International Conference on Image Analysis and Processing, 2007,
• Q. Zhang, K. Chandramouli, U. Damnjanovic, T. Piatrik, E. Izquierdo, M. Corvaglia, N. Adami, R. Leonardi, G. Yakin, S. Aksoy, U. Naci, A. Hanjalic, S. Vrochidis, A. Moumtzidou, S. Nikolopoulos, V. Mezaris, L. Makris, I. Kompatsiaris, B. Mansencal, J. Benois-Pineau, E. Esen, A. A. Alatan, E. Spyrou, P. Kapsalas, G. Tolias, P. Mylonas, Y. Avrithis, B. Reljin, G. Zajic, A. M. G. Pinherio, L. A. Alexandre, P. Almeida, R. Jarina, M. Kuba, N. Aginako, and J. Goya, "The COST292 experimental framework for TRECVID 2007," in Proc. TRECVid 2007 - Text Retrieval Conference TRECVid Workshop, 2007,
• U. Damnjanovic, T. Piatrik, D. Djordjevic, and E. Izquierdo, "Video Summarisation for Surveillance and News Domain," in Proc. International conference on Semantics and digital Media Technologies, 2007,
• T. Piatrik and E. Izquierdo, "Semantic Multimedia Retrieval using Ant Colony Inspired Methods," in Proc. K-Space Jamboree Workshop, 2007,
• P. Wilkins, T. Adamek, P. Ferguson, M. Hughes, G. Jones, G. Keenan, K. McGuinness, J. Malobabic, N. O'Connor, D. Sadlier, A. F. Smeaton, R. Benmokhtar, E. Dumont, B. Huet, B. Merialdo, E. Spyrou, G. Koumoulos, Y. Avrithis, R. Moerzinger, P. Schallauer, W. Bailer, Q. Zhang, T. Piatrik, K. Chandramouli, E. Izquierdo, L. Goldmann, M. Haller, T. Sikora, P. Praks, J. Urban, X. Hilaire, and J. M. Jose, "K-Space at TRECVid 2006," in Proc. TRECVid 2006 - Text Retrieval Conference TRECVid Workshop, 2006,
• T. Piatrik, K. Chandramouli, and E. Izquierdo, "Image Classification using Biologically Inspired Systems," in Proc. 2nd International Mobile Multimedia Communications Conference, 2006,
• T. Piatrik and E. Izquierdo, "Image Classification using an Ant Colony Optimization approach," in Proc. 1st International Conference on Semantic and Digital Media Technologies, 2006,

Technical reports

• K. Chandramouli, T. Piatrik, T. Kliegr, J. Nemrava, and E. Izquierdo, "Final report on Biologically Inspired K-Space Classifiers," Queen Mary, University of London, 2008.