Vous êtes ici : Réunions » Réunion

Identification

Identifiant: 
Mot de passe : 

Mot de passe oublié ?
Détails d'identification oubliés ?

Fingerprint - Action 2 Protection des données visuelles

Nous vous rappelons que, afin de garantir l'accès de tous les inscrits aux salles de réunion, l'inscription aux réunions est gratuite mais obligatoire.

Inscriptions closes à cette réunion.

Inscriptions

10 personnes membres du GdR ISIS, et 0 personnes non membres du GdR, sont inscrits à cette réunion.
Capacité de la salle : 50 personnes.

Annonce

Thème D - Action 2 "Protection des données visuelles"
Journée "Fingerprint"

Lieu : Jardins de l'Innovation, à Orange Labs Issy-les-Moulineaux
France Télécom R&D, 38 rue du Général Leclerc, 92130 Issy-les-Moulineaux
Depuis Montparnasse : métro 12 direction Mairie d'Issy, arrêt Corentin Celton

Date : Vendredi 1er Juillet 2011.
Journée spéciale "Fingerprint"
Personne invitée : Gábor TARDOS,  Alfréd Rényi Institute of Mathematics (Hungarian Academy of Sciences) and School of Computing Science (Simon Fraser University)

Objectifs :

Depuis 10 ans, la protection des données numériques prend de plus en plus d'importance. Ceci vient du fait que nous créons, transmettons et archivons de grandes quantités de données multimédia et en particuliers, des musiques, des images, des vidéos ainsi que des scènes 3D. La protection de ces données multimédia peut se faire par l'intermédiaire de techniques permettant de tracer ces données comme le fingerprinting.

Le terme fingerprint a des multiples définitions riches et variées allant de la création d'empreintes ou signatures à la traçabilité de documents, en passant par la mise en place de protocoles. L'objectif de cette journée sera de faire le point sur les dernières avancées du "Fingerprint" afin d'en présenter ses différentes facettes. Les thèmes abordés seront :

Contact :

Patrick Bas, LAGIS , Lille [patrick.bas@ec-lille.fr]
Gaëtan LE GUELVOUIT, Orange-Labs, Rennes [gaetan.leguelvouit@orange-ftgroup.com]
et William Puech, ICAR/LIRMM, Montpellier [william.puech@lirmm.fr].

plan d'acces

Programme

Cette journée sera composée, le matin, de trois exposés longs dont celui de Gabor TARDOS et l'après-midi de présentations plus courtes des activités Françaises dans ce domaine. L'organisation sera commune entre le GdR et Orange Labs.

Résumés des contributions

Trading length for efficiency in digital fringerprints

Invited talk.
Gábor TARDOS, Alfréd Rényi Institute of Mathematics (Hungarian Academy of Sciences) and School of Computing Science (Simon Fraser University)

Towards joint decoding of Tardos fingerprinting codes

Teddy FURON, INRIA Rennes, France

The class of joint decoder of probabilistic fingerprinting codes isof utmost importance in theoretical papers to establish the concept of fingerprint capacity [Huang:2010,Amiri:2009,Anthapadmanabhan:2008]. However, no practical implementation supporting a large user base is known to date. This paper presents an iterative decoder which is, as far as we are aware of, the first practical attempt towards joint decoding. The discriminative feature of the scores benefits on one hand from the side-information of previously accused users, and on the other hand, from recently introduced universal linear decoders for compound channels [Abbe:2010]. The probability of accusing innocents is under control thanks to rare event simulation. Neither the code construction nor the decoder make precise assumptions about the collusion (size or strategy). An intensive experimental work benchmarks the very good performances and offers a clear comparison with previous state-of-the-art decoders.

Approximate search as a source coding problem, with application to large scale image retrieval

Hervé JEGOU, INRIA Rennes, France.

Traitor tracing for the multimedia entertainment industry

Gwenaël DOERR, Technicolor, Rennes, France.

Considering Security and Robustness Constraints for Watermark-based Tardos Fingerprinting

Benjamin MATHON, Gipsa-Lab, Grenoble, France.

This talk deals with a theoretical study onbinary Tardos' fingerprinting codes embedded using watermarking schemes. Our approach is derived from recent works from Furon et. al. and encompasses both security and robustness constraints. We assume here that the collusion has estimated the symbols of the fingerprinting code by the way of a security attack, the quality of the estimation relying on the security of the watermarking scheme. Taking into account the fact that the collusion can perform estimation errors, we update the Worst Case Attack, which minimizes the mutual information between the sequence of one colluder and the pirated sequence forged by the collusion. After comparing the achievable rates of the previous and proposed Worst Case Attack according to the estimation error,  we conclude this analysis by comparing the robustness of no-secure embedding schemes versus secure ones. We show that, for low probabilities of error during the decoding stage (e.g. highly robust watermarking schemes), security enables to increase the achievable rate of the fingerprinting scheme.

Probabilist Fingerprinting Codes Used To Detect Traitor's Zero-bit Watermark

Mathieu DESOUBEAUX, Orange Labs, Rennes, LIRMM, Montpellier, France.

Securing H.264 video with Tardos fingerprinting code using spread spectrum embedding

Zafar SHAHID, IRCCyN, Nantes, France.

Private anonymous fingerprinting for color image in the wavelet domain

Wadood ABDUL, Philippe GABORIT and Philippe CARRE, XLIM, Poitiers, Limoges, France.

In private anonymous fingerprinting, an online buyer of multimedia content does not want to reveal his identity or his choice of multimedia content, whereas the seller or owner of the content does not want the buyer to further distribute the content illegally. We present a new private anonymous fingerprinting protocol. It is based on superposed sending for communication security, group signature for anonymity and traceability and single database private information retrieval (PIR) to allow the user to get an element of the database without giving any information about the acquired element.

Asymmetric fingerprinting protocols

Ana CHARPENTIER, Caroline FONTAINE, Teddy FURON et Ingemar COX, INRIA, Rennes, Telecom Bretagne, Brest, France.

Asymmetric fingerprinting protocols are designed to prevent an untrustworthy Provider incriminating an innocent Buyer. These protocols enable the Buyer to generate his/her own fingerprint by him/herself, and ensure that the Provider never has access to the Buyer's copy of the Work. Until now, such protocols were not easy to use in practice, because the collusion-resistant codes they rely on were too long to be practical.  But now, with the short and powerful Tardos probabilistic collusion-resistant codes, things have changed. However, previous asymmetric fingerprinting protocols cannot be directly applied to Tardos codes, because the fingerprint generation depends on a secret vector that is only known to the Provider.  Here we propose a new asymmetric fingerprinting protocol, specifically for Tardos codes, that solves this issue. We also take into account a new cheating strategy for the Provider, which is also specific to Tardos codes, as he/she may use a false secret vector to incriminate any innocent Buyer, without knowing the Buyer's copy of the Work.

Date : 2011-07-01

Lieu : Jardins de l'Innovation, à Orange-Labs Issy-les-Moulineaux

Thèmes scientifiques :
D - Télécommunications : compression, protection, transmission

Inscriptions closes à cette réunion.

Accéder au compte-rendu de cette réunion.

(c) GdR IASIS - CNRS - 2024.