Stamatios Lefkimmiatis is an Assistant Professor at Skolkovo Institute of Science and Technology (Skoltech). He received the Diploma degree in Computer Engineering & Informatics from University of Patras, Greece in 2004 and the Ph.D degree in Electrical and Computer Engineering from the National Technical University of Athens (NTUA), Greece in 2009. Prior to joining Skoltech, he held research positions in Ecole Polytechnique Fédérale de Lausanne (EPFL), Switzerland and University of California, Los Angeles (UCLA) , CA, USA.
His research interests lie in the areas of image analysis, machine learning, and computer vision. His current focus is on inverse
problems in imaging, such as denoising, deconvolution, interpolation, super-resolution, inpainting, and sparse reconstruction, with
applications in digital photography, biomicroscopy, remote sensing, medical and astronomical imaging.
Journal Articles A. Saucedo, S. Lefkimmiatis, R. Novena, and K. Sung, “Improved Computational Efficiency of Locally Low Rank MRI Reconstruction Using Iterative Random Patch Adjustments”, IEEE Trans. Medical Imaging, vol. 36, issue 6, pp. 1209-1220, June 2017 [pdf][appendix]
 S. Lefkimmiatis, “Universal Denoising Networks : A Novel CNN Architecture for Image Denoising,” IEEE Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA, June 2018. [pdf][appendix][software] (New)
 S. Lefkimmiatis, “Non-Local Color Image Denoising with Convolutional Neural Networks,” IEEE Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, July 2017. [pdf][appendix][poster][software]
 J. Lin, S. Lefkimmiatis, and S. Kyunghyun,“Deep network training based sparsity model for reconstruction”, In Proc. 25th ISMRM Meeting, Hawaii, USA, April 2017.
 S. Lefkimmiatis, A. Saucedo, S. Osher, and S. Kyunghyun,“Vectorial non-local total variation regularization for calibration-free parallel MRI reconstruction”, In Proc. Int. Symp. Biomedical Imaging (ISBI’15), Brooklyn, NY, Apr. 2015. [pdf]
 A. Saucedo, S. Lefkimmiatis, S. Osher, and S. Kyunghyun,“Novel non-local total variation regularization for constrained MR reconstruction”, In Proc. 23rd ISMRM Meeting, Toronto, Canada, May 2015. [pdf]
 S. Kromwijk, S. Lefkimmiatis and M. Unser, “High-performance 3D Deconvolution of Fluorescence Micrographs,” In Proc. Int. Conf. Image Processing (ICIP’14), Paris, France, Oct. 2014, pp. 3029–3032. [pdf]
 E. Froustey, E. Bostan, S. Lefkimmiatis, and M. Unser, “Digital phase reconstruction via iterative solutions of transport-of-intensity equation,” in Proc. 13th IEEE Workshop on Information Optics (WIO’14), Neuchatel NE, Switzerland, July 2014, pp. 1–3. [pdf]
 S. Lefkimmiatis, A. Roussos, M. Unser, and P. Maragos, “Convex Generalizations of Total Variation based on the Structure Tensor with Applications to Inverse Problems,” In Scale Space and Variational Methods in Computer Vision, Springer Berlin Heidelberg, 2013, vol. 7893, pp. 48–60. [pdf] [appendix]
 S. Lefkimmiatis and M. Unser, “3D Poisson Microscopy Deconvolution with Hessian Schatten-Norm Regularization,” In Proc. Int. Symp. Biomedical Imaging (ISBI’13), San Francisco, CA, Apr. 2013, pp. 165–168. [pdf]
 D. Sage, H. Kirshner, C. Vonech, S. Lefkimmiatis, and M. Unser, “Benchmarking Image-Processing Algorithms for Biomicroscopy: Reference Datasets and Perspectives,” In Proc. 21st European Signal Processing Conf. (EUSIPCO’13), Marrakech, Morocco, Sep. 2013.
 S. Lefkimmiatis and M. Unser, “A Projected Gradient Algorithm for Image Restoration under Hessian Matrix-Norm Regularization,” In Proc. Int. Conf. Image Processing (ICIP’12), Orlando, FL, Sep. 2012, pp. 3029–3032. [pdf] [software]
 S. Lefkimmiatis, A. Bourquard, and M. Unser, “Hessian-Based Regularization For 3-D Microscopy Image Restoration,” In Proc. Int. Symp. Biomedical Imaging (ISBI’12), Barcelona, Spain, May 2012, pp. 1731–1734. [pdf]
 D. Zafer, S. Lefkimmiatis, A. Bourquard, and M. Unser, “A Second-order Extension of TV Regularization for Image Deblurring,” In Proc. Int. Conf. Image Processing (ICIP’11), Brussels, Belgium, Sep. 2011, pp. 713–716.
 S. Lefkimmiatis, G. Papandreou, and P. Maragos, “Poisson-Haar transform: A nonlinear multiscale representation for photon-limited image denoising”, In Proc. IEEE Int. Conf. Image Processing (ICIP’09), Cairo, Egypt, Nov. 2009, pp. 3853–3856. [pdf]
 S. Lefkimmiatis, G. Papandreou, and P. Maragos, “Photon-limited image denoising by inference on multiscale models,” In Proc. IEEE Int. Conf. Image Processing (ICIP’08), San Diego, CA, Oct. 2008, pp. 2332–2335. [pdf]
 S. Lefkimmiatis, P. Maragos, and A. Katsamanis, “Multisensor multiband cross-energy tracking for feature extraction and recognition”, In Proc. IEEE Int. Conf. Acoustics, Speech, and Signal Processing (ICASSP’08), Las Vegas, NV, Apr. 2008, pp. 4741–4744. [pdf]
 D. Dimitriadis, P. Maragos, and S. Lefkimmiatis, “Multiband, multisensor features for robust speech recognition,” In Proc. Int. Conf. Speech Technology and Communication (ICSTC’07), Antwerp, Belgium, Aug. 2007. [pdf]
 S. Lefkimmiatis and P. Maragos, “Optimum Post-Filter Estimation for Noise Reduction in Multichannel Speech Processing,” In Proc. 14th European Signal Processing Conf. (EUSIPCO’06), Florence, Italy, Sep. 2006. [pdf]
 S. Lefkimmiatis, D. Dimitriadis, and P. Maragos, “An Optimum Microphone Array Post-Filter for Speech Applications,” In Proc. Int. Conf. Spoken Language Processing (ICSLP’06), Pittsburgh, PA, Sep. 2006, pp 2142–2145. [pdf]
 F. Kokkinos and S. Lefkimmiatis, “Deep Image Demosaicking using a Cascade of Convolutional Residual Denoising Networks,” March 2018. [pdf]
 S. Lefkimmiatis, “Universal Denoising Networks : A Novel CNN-based Network Architecture for Image Denoising,” Nov. 2017. [pdf]
 S. Lefkimmiatis, “Non-Local Color Image Denoising with Convolutional Neural Networks,” Nov. 2016. [pdf]
Ph.D positions are currently available at the Computation Imaging Group at Skoltech in the areas of computer vision and image processing. The research focus will be on mathematical image modeling, computational methods for inverse imaging problems, large-scale optimization, and fast numerical methods.
Successful candidates are expected to have a masters degree in electrical engineering, computer science, or applied mathematics. Previous experience in image/signal processing and/or machine learning is highly desirable. Applicants should have a strong theoretical background and a desire to use mathematical tools in their research. A good knowledge of C/C++, Python, Matlab and/or Cuda is also required.
Students interested in joining the Computational Imaging Group for a Ph.D. thesis are advised to first contact Prof. Stamatios Lefkimmiatis. The standard procedure is to apply via Skoltech Admissions for Doctoral Programs.
ФИО: Стаматиос Лефкиммиатис
Занимаемая должность (должности): Старший преподаватель
Преподаваемые дисциплины: –
Ученая степень: PhD, проектирование электрических систем и вычислительной техники, 2009 г., Афинский национальный технический университет
Ученое звание: нет
Наименование направления подготовки и/или специальности: вычислительная техника и информатика
Данные о повышении квалификации и/или профессиональной переподготовке: нет
Общий стаж работы: 7 лет
Стаж работы по специальности: 7 лет