George Retsinas

Intelligent Robotics and Automation Laboratory, National Technical University of Athens

prof_pic.jpg

I am George Retsinas, first and foremost a problem solver and machine learning enthusiast, currently working as a Postdoctoral Researcher in the School of Electrical and Computer Engineering of the National Technical University of Athens, where I have received my Ph.D. focused on machine learning and computer vision. I have also been a research associate at Athena Research Center and National Center for Scientific Research “Demokritos”. The main topic of research during my Ph.D. was Document Analysis and Recognition and still remains one of my side interests, while the main theme of my current research is robotic vision. My main passion revolves around deep learning optimization (e.g., model compression, self-supervised learning), trying to achieve the balance between theoretical research and applied science.

news

Apr 4, 2024 SMIRK has been accepted at CVPR 2024 !

selected publications

  1. 3D Facial Expressions through Analysis-by-Neural-Synthesis
    G. Retsinas, P. P. Filntisis, R. Danecek, V. F. Abrevaya, A. Roussos, T. Bolkart, and P. Maragos
    In Conference on Computer Vision and Pattern Recognition (CVPR), 2024
  2. Feather: An Elegant Solution to Effective DNN Sparsification
    A.G. Georgoulakis, G. Retsinas, and P. Maragos
    In British Machine Vision Conference (BMVC), 2023
  3. Neural Network Approximation based on Hausdorff distance of Tropical Zonotopes
    P. Misiakos, G. Smyrnis, G. Retsinas, and P. Maragos
    In International Conference on Learning Representations (ICLR), 2022
  4. From Seq2Seq Recognition to Handwritten Word Embeddings
    G. Retsinas, G. Sfikas, C. Nikou, and P. Maragos
    In British Machine Vision Conference (BMVC), 2021
  5. An Alternative Deep Feature Approach to Line Level Keyword Spotting
    G. Retsinas, G. Louloudis, N. Stamatopoulos, G. Sfikas, and B. Gatos
    In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019
  6. RecNets: Depthwise-Recurrent Convolutional Neural Networks
    G. Retsinas, A. Elafrou, G. Goumas, and P. Maragos
    In British Machine Vision Conference (BMVC), 2019
  7. Efficient Learning-Free Keyword Spotting
    G. Retsinas, G. Louloudis, N. Stamatopoulos, and B. Gatos
    IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018