Section: Partnerships and Cooperations

National Initiatives

ANR-PERSEE

Participants : Josselin Gautier, Christine Guillemot, Laurent Guillo, Olivier Le Meur, Fabien Racapé.

• Title  : Perceptual coding for 2D and 3D images.

• Research axis : §  6.2.2 ,  6.1.1 .

• Partners : IRCCYN-Polytech Nantes, INSA-Rennes, Telecom Paris Tech.

• Funding : ANR.

• Period : 10/2009-08/2013

The objective of the project is to develop perceptually driven coding solutions for mono-view and multi-view video. The SIROCCO project-team contributes on different problems relevant for mono-view and multi-view video coding: visual attention modeling (see Section 6.1.1 ), on texture synthesis and inpainting for both 2D and 3D content. Several methods for 2D image inpainting and 2D/3D inpainting to handle disocclusions in virtual view synthesis have been developed (see Sections 6.2.2 . A computational model for 3D content has also been studied (see Section 6.1.1 )

ANR-ARSSO

Participants : Mounira Ebdelli, Christine Guillemot, Ronan Le Boulch, Olivier Le Meur, Aline Roumy.

• Title  : Adaptable, Robust, Streaming SOlutions.

• Partners : Inria/Planète, TESA-ISAE, CEA-LETI/LNCA, ALCATEL LUCENT BELL LABS, THALES Communications, EUTELSAT SA.

• Funding : ANR.

• Period : 06/2010-11/2013

The ARSSO project focuses on multimedia content communication systems, characterized by more or less strict real-time communication constraints, within highly heterogeneous networks, and toward terminals potentially heterogeneous too. It follows that the transmission quality can largely differ in time and space. The solutions considered by the ARSSO project must therefore integrate robustness and dynamic adaptation mechanisms to cope with these features. The overall goal is to provide new algorithms, develop new streaming solutions and study their performances. The SIROCCO project-team contributes on the development of loss concealment methods based on video inpainting. A first approach using examplar-based inpainting with neighbor embedding techniques has been developed. This method is currently being improved along three directions: 1/- he use of new distance metrics for finding the best matching patches; 2/- using a multi-resolution approach to both reduce the computational time and improve the robustness of the method; 3/- using mosaicking techniques for enhancing steps of stationary background and spatial inpainting. These solutions are studied in the context of a video compression and transmission chain using the emerging HEVC coding standard.