about me

I am a research associate at Centre Borelli (ex CMLA), at ENS Paris-Saclay (ex ENS Cachan). My work focuses in image and video restoration using machine leaning and image processing.

Before that I worked as a researcher in Fundació Barcelona Media (a private non-profit research center, now absorbed by Eurecat). Back then I worked in video processing for visual effects in movie post-production, and in real-time RGB+D video processing for virtual broadcast studios.

I did my Ph.D. at the University Pompeu Fabra in Barcelona, Spain, under the supervision of Prof. Vicent Caselles on image inpainting and video editing.

I am from Uruguay, where in 2006 I graduated as a Telecommunications Engineer from the Facultad de Ingeniería (Engineering School) of the Universidad de la República.
knobs















research

Papers
 
Self-supervised multi-image super-resolution for push-frame satellite images
Ngoc Long Nguyen, Jérémy Anger, Axel Davy, Pablo Arias, Gabriele Facciolo
Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). Earth Vision Workshop, 2021
 
Proba-V-ref: Repurposing the Proba-V challenge for reference-aware super resolution
Ngoc Long Nguyen, Jérémy Anger, Axel Davy, Pablo Arias, Gabriele Facciolo
2021 IEEE International Geoscience and Remote Sensing Symposium (IGARSS)
 
Self-Supervised Training for Blind Multi-Frame Video Denoising
Valery Dewil, Jeremy Anger, Axel Davy, Thibaud Ehret, Gabriele Facciolo, Pablo Arias
Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2021, pp. 2724-2734
 
Video Denoising by Combining Patch Search and CNNs
Axel Davy, Thibaud Ehret, Jean-Michel Morel, Pablo Arias, Gabriele Facciolo
Journal of Mathematical Imaging and Vision, 63:73–88, 2021
 
Relative Radiometric Normalization Using Several Automatically Chosen Reference Images for Multi-Sensor, Multi-Temporal Series
Charles Hessel, Rafael Grompone von Gioi, Jean-Michel Morel, Gabriele Facciolo, Pablo Arias and Carlo de Franchis
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences (The Annals act as the proceedings of the 2020 edition of the XXIVth ISPRS Congress)
 
A Non-Local CNN for Video Denoising
Axel Davy, Thibaud Ehret, Jean-Michel Morel, Pablo Arias, and Gabriele Facciolo
IEEE Int. Conf. on Image Processing (ICIP), 2019
 
Joint demosaicing and denoising by overfitting of bursts of raw images
Thibaud Ehret, Axel Davy, Pablo Arias and Gabriele Facciolo
Intl. Conf. on Computer Vision (ICCV). Seoul, 2019
 
Kalman filtering of patches for frame recursive video denoising
Pablo Arias and Jean-Michel Morel
IEEE Conf. on Compuper Vision and Pattern Recogition Workshops (CVPRW). NTIRE: New Trends on Image Restoration and Enhancement Workshop. Long Beach, 2019
 
Model-blind Video Denoising Via Frame-to-frame Training
Thibaud Ehret, Axel Davy, Jean-Michel Morel, Gabriele Facciolo and Pablo Arias
IEEE Conf. on Computer Vision and Pattern Recognition (CVPR 2019). Long Beach, 2019
 
Non-Local Kalman: A recursive video denoising algorithm
Thibaud Ehret, Pablo Arias and Jean-Michel Morel
IEEE Int. Conf. on Image Processing (ICIP 2018). Athens, 2018
 
A comparison of Patch-Based Models in Video Denoising.
Pablo Arias, Gabriele Facciolo and Jean-Michel Morel
IEEE 13th Image, Video and Multidimensional Signal Processing Workshop (IVMSP 2018). Aristi, 2018
 
On the convergence of PatchMatch
Thibaud Ehret and Pablo Arias
IEEE Conf. on Computer Vision and Pattern Recognition (CVPR 2018). Salt Lake City, 2018
 
Video denoising via Empirical Bayesian Estimation of Space-Time Patches
Pablo Arias and Jean-Michel Morel
Journal of Mathematical Imaging and Vision, 60(1):70-93, 2017
 
Below the surface of the Non-Local Bayesian Image Denoising method
Pablo Arias and Mila Nikolova
6th Int. Conf. on Scale Space and Variational Methods (SSVM 2017). Kolding, 2017
 
Global patch search boost video denoising.
Thibaud Ehret, Pablo Arias and Jean-Michel Morel.
Proc. of the 12th Int. Joint Conf. on Computer Vision Imaging and Computer Grahpics Theory and Applications (VISIGRAPP 2016), Volume 4: VISAPP. Porto, 2016
 
Visual music transcription of clarinet video recordings trained with audio-based labelled data
Pablo Zinemanas, Pablo Arias, Gloria Haro and Emilia Gómez
Int. Conf. on Computer Vision Workshop on Computer Vision for Audio-Visual Media (CVAVM). Venice, 2017
 
Exemplar-based image inpainting using an affine invariant similarity measure
Vadim Fedorov, Pablo Arias, Gabriele Facciolo and Coloma Ballester
VISIGRAPP 2016 (Revised Selected Papers)
 
Affine invariant self-similarity for exemplar-based inpainting.
Vadim Fedorov, Pablo Arias, Gabriele Facciolo and Coloma Ballester.
Proceedings of the 11th Int. Joint Conf. on Computer Vision Imaging and Computer Grahpics Theory and Applications (VISIGRAPP), Volume 3: VISAPP. Rome, 2016
 
Variational framework for non-local inpainting
Vadim Fedorov, Gabriele Facciolo and Pablo Arias
Image Processing On Line 5:362-386, 2015
 
Linear multiscale analysis of similarities between images on Riemannian manifolds: Practical formula and affine covariant metrics
Vadim Fedorov, Pablo Arias, Rida Sadek, Gabriele Facciolo, Coloma Ballester
SIAM Journal on Imaging Sciences 8(3):2021-2069, 2015
 
Towards a Bayesian video denoising method
Pablo Arias and Jean-Michel Morel
Int. Conf. on Advanced Concepts for Intelligent Vision Systems (ACIVS). Catania, 2015
 
A Variational Model for Gradient-Based Video Editing
Rida Sadek, Gabriele Facciolo, Pablo Arias and Vicent Caselles
International Journal of Computer Vision, 103(1):127-162, 2012
 
Nonlocal variational models for inpainting and interpolation
Pablo Arias, Vicent Caselles, Gabriele Facciolo, Vanel Lazcano and Rida Sadek
Mathematical Models and Methods in Applied Sciences, 22, 2012
 
Analysis of a Variational Framework for Exemplar-Based Image Inpainting
Pablo Arias, Vicent Caselles and Gabriele Facciolo
Multiscale Modeling and Simulation, 10(2), 2012
 
A gradient-based neighborhood filter for disparity interpolation
Vanel Lazcano, Pablo Arias, Gabriele Facciolo and Vicent Caselles
IEEE Int. Conf. on Image Processing (ICIP). Miami, 2012
 
A Variational Framework for Exemplar-Based Image Inpainting
Pablo Arias, Gabriele Facciolo, Vicent Caselles and Guillermo Sapiro
International Journal of Computer Vision, 93(3):319-347, 2011
 
Exemplar-based interpolation of sparsely sampled images
Gabriele Facciolo, Pablo Arias, Vicent Caselles and Guillermo Sapiro
7th Int. Conf. on Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR). Bonn, 2009
 
A variational framework for non-local image inpainting
Pablo Arias, Vicent Caselles and Guillermo Sapiro
7th Int. Conf. on Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR). Bonn, 2009
 
Connecting the Out-of-Sample and the Pre-Image Problems in Kernel Methods
Pablo Arias, Gregory Randall and Guillermo Sapiro
Proceedings of IEEE Conf. on Computer Vision and Pattern Recognition. Minneapolis, 2007
 
Ultrasound Image Segmentation with Shape Priors: Application to Automatic Cattle Rib-Eye Area Estimation
Pablo Arias, Alejandro Pini, Pablo Sprechmann, Gonzalo Sanguinetti, Pablo Cancela, Alicia Fernández, Álvaro Gómez and Gregory Randall
IEEE Transactions on Image Processing, 16(6):1637-1645, 2007
 
Video Analysis Platform
Pablo Flores, Federico Lecumberry, Pablo Arias and Álvaro Pardo
XII Congreso Argentino de Ciencias de la Comunicación, Octubre de 2006.



Patents
 
Method and system for remotely measuring the volume of liquid stored in external floating roof tanks
Carlo De Franchis, Guillaume Lostis, Hefdhi Abdennadher, Pablo Arias, Thomas Madaule, Axel Davy, Sylvain Calisti, Jean-Michel Morel.
US Patent Application No. 20180336693 (2018).
 
Computer-implemented method and system for processing video with temporal consistency.
Juan Francisco Garamendi, Pablo Arias and Rida Sadek.
European Patent Application No. EP3249609A1 (2016).



Ph.D. Dissertation
 
Variational methods for exemplar-based image inpainting and gradient-domain video editing
Pablo Arias
Universitat Pompeu Fabra, October, 2013



















last modified: 19 Apr 2021; heavily based on mauricio delbracio's webpage. css by martín rocamora

Valid XHTML 1.0 Strict