I am Research Engineer at Facebook Reality Labs based in Zurich. I develop technologies for energy-efficient on-device inference to enable AI on AR devices. Previously, I was as a Senior Engineer at GoPro Paris, where I worked for over 5 years. My research interests include neural networks for image and video restoration, computer vision, image processing, and embedded systems. I've also worked on control theory. I obtained a PhD in Applied Mathematics from Paris Descartes University, a master's degree in Applied Mathematics from ENS Cachan, and a master's degree in Electrical Engineering from Universidad de la República.
C. Laroche, A. Almansa, and M. Tassano. Deep Model-Based Super-Resolution with Non-uniform Blur. Winter Conference on Applications of Computer Vision (WACV), 2023
C. Laroche, A. Almansa, and M. Tassano. Bridging the Domain Gap in Real World Super-Resolution. 2022 IEEE International Conference on Image Processing (ICIP), 2022.
C. Laroche, A. Almansa, E. Coupete, and M. Tassano. Provably Convergent Plug & Play Linearized ADMM, applied to Deblurring Spatially Varying Kernels. To appear in 2023 ICASSP Workshop.
A. Monod, J. Delon, M. Tassano, and A. Almansa. Video Restoration with a Deep Plug-and-Play Prior. Arxiv preprint
M. Tassano, J. Delon, and T. Veit. FastDVDnet: Towards Real-Time Deep Video Denoising Without Flow Estimation. CVPR 2020, oral presentation
M. Tassano, J. Delon, and T. Veit. DVDnet: A Fast Network for Deep Video Denoising. In IEEE International Conference on Image Processing, 2019. Paper selected as one of the 10% papers of the conference!
M. Tassano, J. Delon, and T. Veit. DVDnet: Un réseau profond pour le débruitage vidéo. In XXVIIe Colloque GRETSI, 2019
M. Tassano, J. Delon, and T. Veit, An Analysis and Implementation of the FFDNet Image Denoising Method, Image Processing On Line, 9 (2019), pp. 1–25.
M. Tassano, J. Delon, T. Veit, Multiscale Denoising of Raw Images with Precise Estimation of Noise in All Scales, Poster presented at: SIAM Conference on Imaging Science, Bologna, Italy, 2018
M. Tassano, P. Monzon, J. Ramos, G. De Martino and J. Pechiar, An inexpensive Attitude Determination system for the Uruguayan Cubesat, AntelSat,” 2014 IEEE International Instrumentation and Measurement Technology Conference (I2MTC) Proceedings, Montevideo, 2014, pp. 723-727
M. Tassano, P. Monzon and J. Pechiar, Attitude determination and control system of the uruguayan cubesat, AntelSat, 2013 16th International Conference on Advanced Robotics (ICAR), Montevideo, 2013, pp. 1-6.
M. Tassano, G. Sotta, L.I. de León, G. Gutierrez, A UKF-based Attitude Determination Approach. Poster presented at: 7th European CubeSat Symposium; 2015 Sep 9-11; Liège, Belgium
L.I. de León, G. Sotta, M. Tassano, G. Gutierrez, Lessons learned, Subsystem overview and Primary results from AntelSat, first Uruguayan satellite. Poster presented at: 7th European CubeSat Symposium; 2015 Sep 9-11; Liège, Belgium
A. Monod, J. Delon, and M. Tassano. 2022. Apparatus and Methods for Plug-and-Play Video De-Blur US Patent Application Serial No. 63/374,559, filed Sep 2022. Patent Pending
E. Coupeté, M. Tassano, and N. Bessou. Detection Of Hand Obstruction For Image Capture Device US 11308597 B1, United States Patent and Trademark Office, Apr 19, 2022
M. Tassano, J. Delon, and T. Veit. 2019. Burst Deblurring With Kernel Estimation Networks US Patent Application Serial No. 62/911,657, filed August 2019. Patent Pending
M. Tassano, J. Delon, and T. Veit. 2019. A Method and Apparatus for Convolutional Neural Network-Based Video Denoising US Patent Application Serial No. 62/866,885, filed June 2019. Patent Pending
M. Tassano, J. Delon, and T. Veit. A Method and Apparatus for Convolutional Neural Network-Based Video Denoising US 20200364834 A1, United States Patent and Trademark Office, Dec 19, 2020
M. Tassano, J. Delon, and T. Veit. Multiscale Denoising of Videos US 20200151854 A1, United States Patent and Trademark Office, May 14, 2020
M. Tassano, J. Delon, T. Veit, and M. Lebrun. Multiscale Denoising of Raw Images With Noise Estimation US 20190355098 A1, United States Patent and Trademark Office, Nov 21, 2019