Dr. Òscar Divorra

Company:
Telefónica I+D, S.A.U.
Title:
Senior Media Scientist & Manager
Group:
TokBox Inc.
Address:
Diagonal 00 - Plaça d'Ernest Lluch i Martín, 5 - 08019 Barcelona - SPAIN, EU
E-Mail:
oscar@tokbox.com
Mobile/Landline:
+34 93 123 3336
Dr. Òscar Divorra

     Dr. Òscar Divorra has been in the media field for almost fourteen years. He got his M.Sc. in Electrical and Electronics Engineering in 2000 at the Universitat Politècnica de Catalunya (UPC), Barcelona – Spain, and his PhD in Signal Processing Sciences in 2005 at the École Polytéchnique Fédérale de Lausanne (EPFL), Switzerland, where he worked as a researcher in next generation transforms and their application to image and video coding.
     Òscar worked from 2005 util 2008 at Thomson Corporate Research (now Technicolor) in Princeton, USA, leading research on new video modeling strategies for next generation video coding standards, and participated in MPEG and ITU/VCEG forums as Thomson/Technicolor standards delegate. In late 2008, Òscar started collaboration with Telefónica Research (later Telefonica Digital) on next generation Telepresence for more immersive communications. There, he led a multi-company project devoted to explore true eye-contact multi-party 3D Video communications. The project consortium including Philips, Radvision and Fraunhofer-HHI achieved a first-of-a-kind room-class 3D Telepresence system. Later, he further contributed as the technical lead of VoIP projects that helped to push within Telefonica Digital consumer videoconferencing from a digital telco perspective. He joined, on January 2013 Tokbox Inc.. At Tokbox, Òscar is Senior Scientist and has been leading the Intelligent Quality Control / Quality of media Experience engineering initiative. There, he is working on media communications technology around WebRTC for Tokbox platform continuous evolution. He also cares about liaison management matters of Tokbox office in Spain.
    Dr. Divorra has published more than 30 papers in well known journals and conferences and has several patents and patent applications on the video area. He has been recipient of the IEEE ICIP IBM best paper award, and IEEE Signal Processing Society Young Author Best Paper Award. Òscar has been a usual reviewer in several top tier journals and conferences and has been jury member at four international PhD evaluation committees.

J. Civit, O. Divorra Escoda,
Robust Foreground Segmentation for GPU Architechtures in an Immersive 3D Videoconferencing System,
in Proc. of IEEE MMSP, Saint Malo, France, 2010.

O. Divorra Escoda, J. Civit, D. Espinola, R.v. Eijk, W. Ijsselsteijn, F. Zuo, H. Belt, R. Braspenning,
I. Feldmann, O Schreer, W. Waizenneger, E. Yellin, P. Hagendorf,
Towards 3D-Aware Telepresence: Working on Technologies Behind the Scene,
pres. in the Workshop on The Future of Telepresence in ACM CSCW, Savannah, USA, 2010

Y. Zheng, P. Yin, O. Divorra Escoda, J. Solé, C. Gomila,
Prediction Matching for Video Coding,
in Proc. of SPIE VCIP, San Diego, USA, 2010

T. Montserrat, J. Civit, O. Divorra Escoda, J.-L. Landabaso,
Depth Estimation Based on Multiview Matching with Depth/Color
Segmentation and Memory Efficient Belief Propagation,
in Proc. of IEEE ICIP, El Cairo, Egipt, 2009

     Video technologies expert contributing to video processing and imaging and next generation 3D video communications. Mentoring and supervision of R&D engineers. Development of image processing algorithms and systems for next generation 3D video communications, IPR generation and scientific papers publication.

     Responsible for Project Management and Scientific lead of an international Consortium of companies in charge to research, create and develop "a first of a kind", next generation, Telepresence Multi-Codec End-Point system, able of improving immersiveness for conferees in a meeting, and with eye-contact and gesture direction conservation.

 

IEEE ICIP 2010: Reviewer

IEEE 3DTV-CON 2010: Program Committee Member

IEEE ICME 2011 - Workshop on Acoustics and Video Coding and Communications: Technical Committee Member

IEEE 3DTV-CON 2011: Reviewer

IEEE ICIP 2011: Reviewer

Specialties

Telepresence, Signal, Image & Video Processing, Video Coding and Compression, Expert in MPEG-x and H.26x standards, 3D/Multi-view Video, Video communication, Standards Forum Delegate (MPEG / VCEG-ITU), Technical/Scientific Management, Mentor of Junior Researchers & Engineers.

     My main research interest areas concern visual communications, video coding, video processing, 3D data processing and their application for seamless capture and communication of multi-dimensional visual data. I am specially interested on the application of signal processing and mathematical tools such as: multidimensional signal processing, geometry adapted image and video signals approximation, Non-Linear and highly nonlinear signal approximations, matching Pursuits, Basis Pursuits, Transforms, Sparse representations, Wavelets, Signals Modeling, Signal reconstruction from partial data, Signal compression (in particular image and video), Denoising, Deartifacting, Inpainting, Visual communications, 3D visual technologies, multi-view analysis...
     However, my interests extend to the field of signal processing and its applications in a general sense.

My goal is to use all these and more disciplines to solve real-life / application driven technological problems of scientific interest in an efficient, value-added and money-making way.

Journal Publications


O. Divorra Escoda, G. Monaci, R. Figueras i Ventura, P. Vandergheynst and M. Bierlaire, 

Geometric video approximation using weighted matching pursuit,

IEEE Transactions on Image Processing, Vol. 18, Nr. 8, pp. 1703-1716, 2009.


 

G. Monaci, O. Divorra Escoda and P. Vandergheynst,

Analysis of Multimodal Sequences Using Geometric Video Representations,

Signal Processing, Vol. 86, Nr. 12, pp. 3534-3548, 2006.


 

O. Divorra Escoda, L. Granai and P. Vandergheynst,

On the Use of A Priori Information for Sparse Signal Approximations,

IEEE Transactions on Signal Processing, Vol. 54, Nr. 9, pp. 3468-3482, 2006.


 

A. Petrovic, O. Divorra Escoda and P. Vandergheynst,

Multiresolution Segmentation of Natural Images: From linear to Non-Linear Scale-Space Representations,

IEEE Transactions on Image Processing, Vol. 13, Nr. 8, pp. 1104-1114, 2004.

(IEEE Young Author Best Paper Award)

Book Chapters


O. Divorra Escoda,

Toward sparse and geometry adapted video approximations,

Ph.D. Thesis, ITS, École Polytechnique Fédérale de Lausanne (EPFL), Switzerland, 2005.

Conference and Workshop Papers


J. Civit, O. Divorra Escoda, 
Robust Foreground Segmentation for GPU Architechtures in an Immersive 3D Videoconferencing System,
in Proc. of IEEE MMSP, Saint Malo, France, 2010.
 
O. Divorra Escoda, J. Civit, D. Espinola, R.v. Eijk, W. Ijsselsteijn, F. Zuo, H. Belt, R. Braspenning,
I. Feldmann, O Schreer, W. Waizenneger, E. Yellin, P. Hagendorf, 
Towards 3D-Aware Telepresence: Working on Technologies Behind the Scene,
pres. in the Workshop on The Future of Telepresence in ACM CSCW, Savannah, USA, 2010
 
Y. Zheng, P. Yin, O. Divorra Escoda, J. Solé, C. Gomila,
Prediction Matching for Video Coding,
in Proc. of SPIE VCIP, San Diego, USA, 2010
 
I. Feldmann, O. Schreer, P. Kauff, R. Schäfer, Z. Fei, H.J.W. Belt, O. Divorra Escoda,
Immersive Multi-User 3D Video Communication,
in Proc. of IBC, Amsterdam, The Netherlands, 2009
 
T. Montserrat, J. Civit, O. Divorra Escoda, J.-L. Landabaso,
Depth Estimation Based on Multiview Matching with Depth/Color Segmentation and Memory Efficient Belief Propagation,
in Proc. of IEEE ICIP, El Cairo, Egipt, 2009
 
J.-L. Landabaso, J.-C. Pujol-Acolado, T. Montserrat, D. Marimon, J. Civit, O. Divorra Escoda,
A Global Probabilistic Framewok for the Foreground, Background and Shadow Classification Task,
in Proc. of IEEE ICIP, El Cairo, Egipt, 2009
 
N. Atzpadin, J.-C. Pujol-Acolado, J.-L. Landabaso, I. Feldmann, O. Divorra Escoda,
Multi-View Depth Estimation Based on Visual-Hull Enhanced Hybrid Recursive Matching for 3D Video Conference Systems,
in Proc. of IEEE ICIP, El Cairo, Egipt, 2009
 
Y. Zheng, P. Yin, O. Divorra Escoda, X. Li, C. Gomila, 
Illumination Compensated Template Matching Intra Prediction for Video Coding,
in Proc. of IEEE ICIP, San Diego, USA, 2008
 
C. C. Dorea, O. Divorra Escoda, P. Yin, C. Gomila,
A Direction-Adaptive Deartifacting In-Loop Filter for Video Coding,
in Proc. of IEEE ICIP, San Diego, USA, 2008
 
O. Divorra Escoda, P. Yin and C. Gomila, 
Hierarchical B-Frame Results on Geometry-adaptive Block Partitioning,
VCEG-AH16 Proposal, ITU/SG16/Q6/VCEG, Antalya, Turkey, January 2008
 
O. Divorra Escoda, P. Yin, C. Gomila, 
A Multi-Lattice Direction-Adaptive Deartifacting Filter for Image & Video Coding
in Proc. of PCS, Lisboa, Portugal, 2007.
 
S. Har-Noy, O. Divorra Escoda, P. Yin, C. Gomila, T. Nguyen,
Adaptive In-Loop Prediction Refinement For Video Coding,
in Proc. of IEEE MMSP, Crete, Greece, 2007.
 
O. Divorra Escoda, P. Yin and C. Gomila,
Geometry-Adaptive Block Partitioning on B-Frames,
VCEG-AG13 Proposal, ITU/SG16/Q6/VCEG, Shenzhen, China, October 2007
 
C. Dai, O. Divorra Escoda, P. Yin, X. Li, C. Gomila
Geometry-Adaptive Block Partitioning for Intra Prediction in Image & Video Coding,
in Proc. of IEEE ICIP, San Antonio, Texas, September 2007 
 
O. Divorra Escoda, P. Yin and C. Gomila,
Geometry-Adaptive Block Partitioning,
VCEG-AF10 Proposal, ITU/SG16/Q6/VCEG, San Jose, California, April 2007
 
O. Divorra Escoda, P. Yin, C. Dai, X. Li,
Geometry-Adaptive Block Partitioning for Video Coding,
in Proc. of IEEE ICASSP, Honolulu, Hawaii, April 2007
 
O. Divorra Escoda, L. Granai, M. Lemay, J. Molinero Hernandez, P. Vandergheynst and J. Vesin,

Ventricular and Atrial Activity Estimation Through Sparse ECG Signal Decompositions,

in Proc. of IEEE ICASSP, 2006.

O. Divorra Escoda, M. Flierl and P. Vandergheynst,

Intra-Adaptive Motion-Compensated Lifted Wavelets for Video Coding,

in Proc. of EUSIPCO, Antalya, 2005.

G. Monaci, O. Divorra Escoda and P. Vandergheynst,

Analysis of Multimodal Signals Using Redundant Representations ,

in Proc. of IEEE ICIP, Genova, 2005.

(IBM Best Paper Award)

O. Divorra Escoda and P. Vandergheynst,

A Bayesian Approach to Video Expansions on Parametric Over-Complete 2-D Dictionaries,

in Proc. of IEEE MMSP, 2004.

O. Divorra Escoda and P. Vandergheynst,

Locally Temporal Adaptive Transform Scheme for Sub-band Video Coding,

in Proc. of IEEE ICASSP, 2003.

O. Divorra Escoda and P. Vandergheynst,

Video Coding Using a Deformation Compensation Algorithm Based on Adaptive Matching Pursuit Image Decompositions,

in Proc. of IEEE ICIP, 2003.

O. Divorra Escoda, A. Petrovic and P. Vandergheynst,

Segmentation of Natural Images Using Scale-Space Representations: A Linear and a Non-Linear Approach,

in Proc. of EUSIPCO, 2002.

O. Divorra Escoda, R. Figueras i Ventura, E. Debes and T. Ebrahimi,

Influence of a Large Image Watermarking Scheme Parallelization on Possible Attacks,

in Proc. of SPIE's, 46th Annual Meeting on Optical Science and Technology, Vol. 4472, pp. 175-186, 2001.

T. Ebrahimi, Y. Abdeljaoued, R. Figueras i Ventura and O. Divorra Escoda,

MPEG-7 CAMERA,

in Proc. of IEEE ICIP, Vol. 3, pp. 600-603, 2001.

 

-




 

J. Civit and O. Divorra Escoda

Hybrid Segmentation Using Color And Depth Data For Improved Foreground, Background and Shadow Classification,

unpublished - Telefonica Research Technical Report, 2011.

Young Author Best Paper Award
of the IEEE Signal Processing Society 2007.
 
Co-author of the IBM Best Student Paper Award
in the IEEE International Conference in Image Processing (ICIP) 2005.
 
UCLA Grant to attend the
“Multiscale Geometry in Image Processing and Coding” graduate course
at the Institute of Pure and Applied Mathematics (IPAM) at UCLA, Los Angeles, USA (2004).
.
 
Me at linkedin:
 
My former web page at EPFL:
 
My CV
 
© 2014 Telefonica Research and Innovation