Biosketch

Anastasios (Tassos) Roussos was born in Athens, Greece in 1983. He received the Diploma degree in Electrical and Computer Engineering from National Technical University of Athens in November 2005. His diploma thesis, supervised by Prof. Petros Maragos, was in theory and applications of methods based on Partial Differential Equations (PDEs) for denoising, simplification, inpainting and interpolation of general vector-valued images.

During November 2005 to October 2010, Tassos pursued his PhD in the CVSP Group, school of ECE, NTUA, under the supervision of Prof. Petros Maragos, working in the areas of Image Analysis and Computer Vision. His PhD thesis, entitled "Nonlinear Diffusion in Computer Vision and Statistical Shape Models, with Applications in Image Analysis of Articulators of Voiced and Signed Speech", focused on novel nonlinear diffusion and statistical shape models, with main applications to the research of the human speech production system and automatic sign language recognition.

Since October 2010, Tassos is a postdoctoral research assistant at the School of Electronic Engineering and Computer Science, Queen Mary, University of London. He works with Dr. Lourdes Agapito on the ERC HUMANIS project, focusing on the development of Computer Vision algorithms able to recover detailed 3D models of non-rigid and articulated objects in a completely automated way, using only the original footage.

His current research interests include variational and PDE-based methods and statistical shape models for optic flow, image enhancement and segmentation, object tracking and feature extraction. He is interested in various applications such as non-rigid structure from motion, biomedical image processing, automatic sign language recognition, acoustic-to-articulatory speech inversion, articulatory speech synthesis and computational photography.

[Full CV ... ]

Publications

Journal Papers

• A. Roussos, S. Theodorakis, V. Pitsikalis and P. Maragos,
  Dynamic Affine-Invariant Shape-Appearance Handshape Features and Classification in Sign Language Videos ,
  Submitted to the Journal of Machine Learning Research (JMLR) -- Special Topic on Gesture Recognition.
  [demo videos  NEW!]

• R. Garg, A. Roussos, L. Agapito,
  A Variational Approach to Video Registration with Subspace Constraints ,
  Submitted to the International Journal of Computer Vision (IJCV).
  [ground truth & videos]

• A. Roussos and P. Maragos,
  Reversible Interpolation of Vectorial Images by an Anisotropic Diffusion-Projection PDE,
  International Journal of Computer Vision (IJCV), vol. 84, no. 2, pp. 130 -145, Aug. 2009.
  [pdf] [results] [code by P. Getreuer]

Conference Papers

• R. Garg, A. Roussos and L. Agapito,
  Robust Trajectory-space TV-L1 Optical Flow for Non-rigid Sequences,
  Proc. 8th International Conference on Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR), Saint Petersburg, Russia, July 25-27, 2011.
  [pdf] [ground truth & videos]

• A. Roussos and P. Maragos,
  Tensor-Based Image Diffusions Derived from Generalizations of the Total Variation and Beltrami Functionals,
  Proc. IEEE Int'l Conf. on Image Processing (ICIP-10), Hong Kong, Sep. 26-29, 2010.
  [pdf] [presentation]

• A. Roussos, S. Theodorakis, V. Pitsikalis and P. Maragos,
  Affine-Invariant Modeling of Shape-Appearance Images applied on Sign Language Handshape Classification,
  Proc. IEEE Int'l Conf. on Image Processing (ICIP-10), Hong Kong, Sep. 26-29, 2010.
  [pdf]

• A. Roussos, S. Theodorakis, V. Pitsikalis and P. Maragos,
  Hand Tracking and Affine Shape-Appearance Handshape Sub-Units in Continuous Sign Language Recognition,
  Workshop on Sign, Gesture and Activity, 11th European Conference on Computer Vision (ECCV), Crete, Greece, September 2010.
  [pdf]

• A. Roussos, A. Katsamanis, P. Maragos,
  Tongue Tracking in Ultrasound Images with Active Appearance Models,
  Proc. IEEE Int'l Conf. on Image Processing (ICIP-09), Cairo, Egypt, Nov. 7-11, 2009.
  [pdf] [poster]

• A. Katsamanis, A. Roussos, P. Maragos, M. Aron and M.-O. Berger
  Inversion from Audiovisual Speech to Articulatory Information by Exploiting Multimodal Data,
  International Seminar on Speech Production (ISSP 2008), Strasbourg, France, Dec. 2008.
  [pdf] [bib] [presentation]

• M. Aron, A. Roussos, M.-O. Berger, E. Kerrien and P. Maragos,
  Multimodality Acquisition of Articulatory Data and Processing,
  16th European Conference on Signal Processing (EUSIPCO), August 25-29, 2008, Lausanne - Switzerland.
  [pdf]

• A. Roussos and P. Maragos,
  Vector-Valued Image Interpolation by an Anisotropic Diffusion-Projection PDE,
  1st International Conference on Scale Space and Variational Methods in Computer Vision, Ischia, Italy, May 30 - June 2, 2007 ; Lecture Notes in Computer Science 4485, pp. 104-115, Springer-Verlag, 2007.
  Received the Best Student Paper Award of the conference.
  [pdf] [presentation] [results]

Theses

• A. Roussos,
  Nonlinear Diffusion in Computer Vision and Statistical Shape Models, with Applications in Image Analysis of Articulators of Voiced and Signed Speech,
  PhD thesis, ECE School, NTUA, October 2010. Thesis supervisor: P. Maragos.
  [pdf (in greek)]
  [presentation in greek]
  [presentation in english]

• A. Roussos,
  Theory and Applications of Partial Differential Equations in Computer Vision   (in greek),
  Diploma thesis, ECE School, NTUA, October 2005. Thesis supervisor: P. Maragos.
  [zip]