George Retsinas

Research Associate, Ph.D.

George Retsinas was born in 1989, in Sparta, Greece. He received the Diploma degree in Electrical and Computer Engineering from National Technical University of Athens in February 2014. His diploma thesis, supervised by Prof. Petros Maragos, tackled fingerspelling, a subtask of sing language recognition. He received his Ph.D. degree from the Department of Electrical and Computer Engineering of National Technical University of Athens in February 2020. Part of his Ph.D. (2015-2018) was done in collaboration with the Institute of Informatics and Telecommunications of the National Center for Scientific Research “Demokritos”. The main research direction during this collaboration was document analysis and recognition, with emphasis on machine learning approaches. Additional research directions, addressed during his PhD, included bio-signal processing and compression of deep neural networks. His main research interests are in Image Processing and Document Image Analysis, Pattern Recognition, Deep Learning, Algorithms and Mathematical Optimization.

R&D Projects

Bonseyes (2018-2020)

BONSEYES – The Artificial Intelligence Marketplace: Development of a Deep Learning Toolbox which aims to move deep learning research from predictive performance of large scale centralised models to cost-sensitive, distributed, sparse satisficing models that are robust to changes in data environments and real-life dirty data scenarios. Research and exploration of compact architectures of deep neural networks and their trade-off between accuracy and memory/time requirements.

READ (2015-2018)

READ - Recognition and Enrichment of Archival Documents: Design and development of services capable of providing access to archival documents with the support of cutting-edge technology such as Handwritten Text Recognition (HTR) and Keyword Spotting (KWS). Research on time/storage efficient machine learning approaches in order to efficiently handle vast amounts of handwritten documents.

OldDocPro (2014-2015)

Development and exploration of state-of-the-art methodologies for processing and recognizing Greek polytonic typewritten and handwritten documents.

Publications


  • G. Retsinas, "Gesture Recognition of the Greek Sign Languages Alphabet using Rotation Invariant Features", supervised by Prof. Petros Maragos, 2014 (in Greek).
  • G. Retsinas, "Visual Representation Learning for Document Image Recognition", supervised by Prof. Petros Maragos, 2020.

  • G. Retsinas, G. Louloudis, N. Stamatopoulos, B. Gatos, “Efficient Learning-Free Keyword Spotting”, IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018.
  • G. Muhlberger, L. Seaward, M. Terras, S. Ares Oliveira, V. Bosch, M. Bryan, S. Colutto, H. Dejean, M. Diem, S. Fiel, B. Gatos, T. Gruning, A. Greinoecker, G. Hackl, V. Haukkovaara, G. Heyer, L. Hirvonen, T. Hodel, M. Jokinen, P. Kahle, M. Kallio, F. Kaplan, K. Florian, R. Labahn, E.M. Lang, S. Laube, G. Leifert, G. Louloudis, R. McNicholl, J.L. Meunier, J. Michael, E. Muhlbauer, N. Philipp, I. Pratikakis, J. Puigcerver Perez, H. Putz, G. Retsinas, V. Romero, R. Sablatnig, J.A. Sanchez, P. Schofield, G. Sfikas, C. Sieber, N. Stamatopoulos, T. Strauss, T. Terbul, A.H. Toselli, B. Ulreich, M. Villegas, E. Vidal, J. Walcher, M. Weidemann, H. Wurster, K. Zagoris, “Transforming Scholarship in the Archives Through Handwritten Text Recognition: Transkribus as a Case Study", Journal of Documentation, 2019.

  • G. Retsinas, B. Gatos, N. Stamatopoulos and G. Louloudis, “Isolated Character Recognition using Projections of Oriented Gradients”, 13th International Conference on Document Analysis and Recognition (ICDAR'15), pp. 336-340, Nancy, France, 2015.
  • G. Retsinas, B. Gatos, A. Antonacopoulos, G. Louloudis and N. Stamatopoulos, “Historical Typewritten Document Recognition Using Minimal User Interaction”, 3rd International Workshop on Historical Document Imaging and Processing (HIP’15), pp. 31-38, Nancy, France, 2015.
  • G. Retsinas, G. Louloudis, N. Stamatopoulos and B. Gatos, “Efficient Document Image Segmentation Representation by Approximating Minimum-Link Polygons”, 12th Workshop on Document Analysis Systems (DAS'16), pp. 293-298, Santorini, Greece, 2016.
  • G. Retsinas, G. Louloudis, N. Stamatopoulos and B. Gatos, “Keyword Spotting in Handwritten Documents using Projections of Oriented Gradients”, 12th Workshop on Document Analysis Systems (DAS'16), pp. 411-416, Santorini, Greece, 2016.
  • G.Sfikas, G. Retsinas, B.Gatos, “Zoning Aggregated Hypercolumns for Keyword Spotting”, 15th International Conference on Frontiers in Handwriting Recognition (ICFHR’16), Shenzhen, China, 2016.
  • G. Sfikas, G. Retsinas, B. Gatos, “A PHOC Decoder for Lexicon-Free Handwritten Word Recognition”, 14th International Conference on Document Analysis and Recognition (ICDAR'17), pp. 513-518, Kyoto, Japan, 2017.
  • G. Retsinas, N. Stamatopoulos, G. Louloudis, G. Sfikas, B. Gatos, “Nonlinear Manifold Embedding on Keyword Spotting using t-SNE”, 14th International Conference on Document Analysis and Recognition (ICDAR'17), pp. 487-492, Kyoto, Japan, 2017.
  • G. Retsinas, G. Sfikas, and B. Gatos, “Transferable deep features for keyword spotting,” in IWCIM, in conjuction with EUSIPCO, 2017.
  • G. Retsinas, Giorgos Sfikas, Nikolaos Stamatopoulos, Georgios Louloudis, Basilis Gatos, “Exploring Critical Aspects of CNN-based Keyword Spotting. A PHOCNet Study”, 13th Workshop on Document Analysis Systems (DAS'18), pp. 13-18, Vienna, Austria, 2018.
  • G. Retsinas, Giorgos Sfikas, Georgios Louloudis, Nikolaos Stamatopoulos and Basilis Gatos, “Compact Deep Decriptors for Keyword Spotting”, 16th International Conference on Frontiers in Handwriting Recognition (ICFHR’18), Niagara Falls, USA, 2018.
  • G. Retsinas, G. Louloudis, N. Stamatopoulos, G. Sfikas and B. Gatos. "An Alternative Deep Feature Approach to Line Level Keyword Spotting." IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 12658-12666, Long Beach, USA, 2019.
  • G. Retsinas, A. Elafrou, G. Goumas and P. Maragos. "RecNets: Channel-wise Recurrent Convolutional Neural Networks.", British Machine Vision Conference (BMVC), Cardiff, UK, 2019.
  • G. Retsinas, P. P. Filntisis, N. Efthymiou, E. Theodosis, A. Zlatintsi and P. Maragos, “Person Identification using Deep Convolutional Neural Networks on Short-Term Signals from Wearable Sensors”, In Proc. of the 45th International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Barcelona, Spain, 2020.
  • G. Smyrnis, G. Retsinas, P. Maragos, “Maxpolynomial Division with Application to Neural Network Simplification”, In Proc. of the 45th International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Barcelona, Spain, 2020.

  • G. Retsinas, A. Elafrou, G. Goumas and P. Maragos, "Weight Pruning via Adaptive Sparsity Loss", arXiv:2006.02768, 2020.

Contact Me

  • Department of Electrical and Computer Engineering,
    National Technical University of Athens,
    Athens, Greece.
  • Phone: +30-210-7722964
  • Email: gretsinas at central.ntua.gr