HAL - Lille 3
Not a member yet
5967 research outputs found
Sort by
Living Labs : de la recherche-action à la réforme de l’action publique ? Le cas du programme d’investissements d’avenir « jeunesse » de la métropole européenne de Lille (France)
International audienc
Etude de l’apport de la reconstruction des régions occultées du visage pour la reconnaissance des expressions
National audienceLa reconnaissance des expressions faciales en contexte d'interaction non contrôlée est, encore aujourd'hui, confrontée à un certain nombre de défis qui réduisent considérablement les performances des systèmes. Parmi ces défis, les occultations induites par les variations de pose de la tête ou par des objets externes (accessoires, mains, ...) impliquent une perte d'informations sur le visage. Différentes méthodes proposées dans la littérature tendent à résoudre ces problèmes soit en reconstruisant les régions occultées, soit en exploitant uniquement les informations au sein des régions visibles du visage. Dans cet article, nous étudions l'apport de la reconstruction des régions occultées, par rapport à l'exploitation efficace de l'information contenue dans les régions visibles du visage lors de la reconnaissance des expressions faciales
Asymptotics of Cholesky GARCH models and time-varying conditional betas
Corresponding publicationsWorking Paper | Asymptotics of Cholesky GARCH Models and Time-Varying Conditional Betas.〈halshs-01944656〉International audienceThis paper proposes a new model with time-varying slope coefficients. Our model, called CHAR, is a Cholesky-GARCH model, based on the Cholesky decomposition of the conditional variance matrix introduced by Pourahmadi (1999) in the context of longitudinal data. We derive stationarity and invertibility conditions and prove consistency and asymptotic normality of the Full and equation-by-equation QML estimators of this model. We then show that this class of models is useful to estimate conditional betas and compare it to the approach proposed by Engle (2016). Finally, we use real data in a portfolio and risk management exercise. We find that the CHAR model outperforms a model with constant betas as well as the dynamic conditional beta model of Engle (2016)
Más et ses « marqueurs du standard » en espagnol : de ~ que.Approche des signifiants en chronosyntaxe interlocutive
International audienceThe question of the distribution de / que as « standard markers » is debated in the literature on comparison in Spanish. In this study, we examine this question in the light of an interlocutive chronosyntax approach, which studies the contribution of signifiers to the dynamics of sense-making by the effets they produce on the interprent at the moment of their occurrence. We propose de and que operate differently in the « receptive framework » of interlocution, and we then show how this receptive framework functions in the various discourse meanings that grammars identify for más de and más que.La question de l'alternance ou de la répartition de / que en position de « marqueur du standard » est l'une des plus débattues de la littérature sur la comparaison en espagnol. Dans cette étude, on se propose de reprendre cette question à la lumière d'une approche en chronosyntaxe interlocutive, qui étudie la contribution des signifiants à la dynamique de construction du sens pour les effets qu'ils produisent sur l'interprétant à l'instant de leur survenance. On propose la formulation d'une contribution opératoire de de et de que en termes de « cadre réceptif » qu'installent ces signifiants dans la relation interlocutive, et on montre ensuite la façon dont ce cadre réceptif sous-tend les divers effets de sens que recensent les grammaires pour más de et más que
MET amplification increases the metastatic spread of EGFR-mutated NSCLC
International audienc
Storyboard-Based Empirical Modelling of Touch Interface Performance
International audienceTouch interactions are now ubiquitous, but few tools are available to help designers quickly prototype touch interfaces and predict their performance. For rapid prototyping, most applications only support visual design. For predictive modelling, tools such as CogTool generate performance predictions but do not represent touch actions natively and do not allow exploration of different usage contexts. To combine the benefits of rapid visual design tools with underlying predictive models, we developed the Storyboard Empirical Modelling tool (StEM) for exploring and predicting user performance with touch interfaces. StEM provides performance models for mainstream touch actions, based on a large corpus of realistic data. We evaluated StEM in an experiment and compared its predictions to empirical times for several scenarios. The study showed that our predictions are accurate (within 7% of empirical values on average), and that StEM correctly predicted differences between alternative designs. Our tool provides new capabilities for exploring and predicting touch performance, even in the early stages of design
Ether-Toolbars: Evaluating Off-Screen Toolbars for Mobile Interaction
International audienceIn mobile touchscreebn interaction, an important challenge is to find solutions to balance the size of individual widgets against the number of widgets needed during interaction. In this work, to address display space limitations, we explore the design of invisible off-screen toolbars (ether-toolbars) that leverage computer vision to expand application features by placing widgets adjacent to the display screen. We show how simple computer vision algorithms can be combined with a natural human ability to estimate physical placing to support highly accurate targeting. Our ether-toolbar design promises targeting accuracy approximating on-screen widget accuracy while significantly expanding the interaction space of mobile devices. Through two experiments, we examine off-screen content placement metaphors and off-screen precision of participants accessing these toolbars. From the data of the second experiment, we provide a basic model that reflects how users perceive mobile surroundings for ether-widgets and validate it. We also demonstrate a prototype system consisting of an inexpensive 3D printed mount for a mirror that supports ether-toolbar implementations. Finally, we discuss the implications of our work and potential design extensions that can increase the usability and the utility of ether-toolbars
Graph sampling with applications to estimating the number of pattern embeddings and the parameters of a statistical relational model
International audienceCounting the number of times a pattern occurs in a database is a fundamental data mining problem. It is a subroutine in a diverse set of tasks ranging from pattern mining to supervised learning and probabilistic model learning. While a pattern and a database can take many forms, this paper focuses on the case where both the pattern and the database are graphs (networks). Unfortunately , in general, the problem of counting graph occurrences is #P-complete. In contrast to earlier work, which focused on exact counting for simple (i.e., very short) patterns, we present a sampling approach for estimating the statistics of larger graph pattern occurrences. We perform an empirical evaluation on synthetic and real-world data that validates the proposed algorithm, illustrates its practical behavior and provides insight into the trade-off between its accuracy of estimation and computational efficiency
Conférence de Cyrille BALLAGUY - Séminaire de Master INALCO - Musée du Quai Branly 2018.: Exposer les mythes gréco-latins : une étude comparative au Parc Astérix et Palais des beaux-arts de Lille.
Docteur en muséologie de l’Université Lille 3, Cyrille Ballaguy est spécialiste de la médiation culturelle. Son champ de recherche s’intéresse à la valorisation de la mythologie gréco-romaine dans les musées du Nord de la France.Dans cette conférence, Cyrille Ballaguy nous expose les mythes gréco-latins, en proposant une étude comparative intéressante au Parc Astérix et Palais des beaux-arts de Lille