Annonce November 2013

annonce@lrde.epita.fr

3 participants
5 discussions

7th European Lisp Symposium, May 5-6 2014, Paris, France

by Didier Verna

ELS'14 - 7th European Lisp Symposium IRCAM, Paris, France May 5-6, 2014 http://www.european-lisp-symposium.org/ The purpose of the European Lisp Symposium is to provide a forum for the discussion and dissemination of all aspects of design, implementation and application of any of the Lisp and Lisp-inspired dialects, including Common Lisp, Scheme, Emacs Lisp, AutoLisp, ISLISP, Dylan, Clojure, ACL2, ECMAScript, Racket, SKILL, Hop and so on. We encourage everyone interested in Lisp to participate. The 7th European Lisp Symposium invites high quality papers about novel research results, insights and lessons learned from practical applications, and educational perspectives. We also encourage submissions about known ideas as long as they are presented in a new setting and/or in a highly elegant way. Topics include but are not limited to: - Context-, aspect-, domain-oriented and generative programming - Macro-, reflective-, meta- and/or rule-based development approaches - Language design and implementation - Language integration, inter-operation and deployment - Development methodologies, support and environments - Educational approaches and perspectives - Experience reports and case studies Please note that IRCAM, the conference venue, is a French institute for research on music and acoustics. Submissions relating Lisp to music or other acoustical matters will hence be particularly welcome, although given no heightened favor during the review process. We invite submissions in the following forms: Papers: Technical papers of up to 8 pages that describe original results or explain known ideas in new and elegant ways. Demonstrations: Abstracts of up to 2 pages for demonstrations of tools, libraries, and applications. Tutorials: Abstracts of up to 4 pages for in-depth presentations about topics of special interest for at least 90 minutes and up to 180 minutes. The symposium will also provide slots for lightning talks, to be registered on-site every day. All submissions should be formatted following the ACM SIGS guidelines and include ACM classification categories and terms. For more information on the submission guidelines and the ACM keywords, see: http://www.acm.org/sigs/publications/proceedings-templates and http://www.acm.org/about/class/1998. Important dates: - TODAY: Mark your calendar. Start planning now! - 09 Mar 2014: Submission deadline - 31 Mar 2014: Notification of acceptance - 21 Apr 2014: Final Papers due - 05 May 2014: Symposium. Join us there! Program Committee: Chair: Kent Pitman, Hypermeta Inc., U.S.A. Local Organizers: Didier Verna, EPITA Research Lab, France Gérard Assayag, IRCAM, France Members: To be announced later Search Keywords: #els2014, ELS 2014, ELS '14, European Lisp Symposium 2014, European Lisp Symposium '14, 7th ELS, 7th European Lisp Symposium, European Lisp Conference 2014, European Lisp Conference '14 -- Resistance is futile. You will be jazzimilated. Lisp, Jazz, Aïkido: http://www.didierverna.info

11 years

Soutenance de thèse de Yongchao Xu

by Yongchao Xu

Bonjour à tous, J'ai le plaisir de vous inviter à ma soutenance de thèse intitulée ``Tree-Based Shape Spaces: Definition and Applications in Image Processing and Computer Vision''. Celle-ci aura lieu le*12 Décembre 2013* à 14h00 dans l'amphithéâtre 260 de l'ESIEE Paris, située au 2, boulevard Blaise Pascal, Cité Descartes, à Noisy-le-Grand (93). Vous trouverez un plan d'accès à l'école à l'adresse suivante : http://www.esiee.fr/Infos-pratiques/acces.php La soutenance sera suivie d'un pot. Composition du jury de thèse ---------------------------- Rapporteurs : Philippe Salembier (Universitat Politècnica de Catalunya) Lionel Moisan (Université Paris Descartes) Examinateurs : Jean Serra (ESIEE Paris) Coloma Ballester (Universitat Pompeu Fabra) Michael H.F. Wilkinson (Rijksuniversiteit Groningen) Beatriz Marcotegui (Mines ParisTech) Pascal Monasse (Ecole des Ponts ParisTech) Directeurs de thèse : Laurent Najman (ESIEE Paris - Université Paris-Est Marne-la-Vallée) Thierry Géraud (EPITA) Résumé de la thèse ------------------ This PhD work presents a versatile framework dealing with tree-based image representations. Those representations have been proved to be very useful in a large number of applications. Notably, in mathematical morphology, those representations form the basis for efficient implementation of connected operators. Connected operators are filtering tools that act by merging some elementary regions of an image. A popular filtering strategy relies on (1) building a tree representation of the image; (2) computing a shape-based attribute on each node of the tree; and (3) removing the nodes for which the attribute is too low. This operation can be seen as a thresholding of the tree, seen as a graph whose nodes are weighted by the attribute function. Rather than being satisfied with a mere thresholding, we propose to expand on this idea. We define the notion of tree-based shape spaces as seeing a tree as a graph whose neighbourhood is given by the parenthood relationship. Then we apply some connected filters on such shape spaces. This simple processing, that we call "shape-based morphology", has several deep consequences. Firstly, it is a generalization of the existing tree-based connected operators. Besides, it allows us to propose two new classes of connected operators: shape-based lower/upper levelings and shapings. Secondly, this framework can be used for object detection/segmentation by selecting relevant nodes in the shape space. Last, we can also apply this framework for transforming hierarchies using extinction values, so that we obtain some hierarchical image simplification and segmentation methods. We have developed several applications using this framework. - A first one is a truly contrast invariant local feature detection method called Tree-Based Morse Regions (TBMR), which can be seen as a variant of MSER with better performances. - Some applications to retinal image analysis (including blood vessel segmentation and optic nerve head segmentation) have been developed using shape-based lower/upper levelings. - We have also proposed an efficient image simplification method by minimizing the Mumford-Shah functional, which belongs to the shaping family. - An object detection scheme have been developed by defining an efficient context-based energy estimator. - Last, we have proposed an extension of constrained connectivity based on our novel hierarchy transforms. Amicalement, Yongchao Xu -- Yongchao XU Ph.D student EPITA Research and Development Laboratory (LRDE) Laboratoire d'informatique de l'Institut Gaspard Monge (LIGM) Équipe Algorithmes, Architectures, Analyse et synthèse d'images (A3SI) Université Paris Est, Marne-la-Vallée France

11 years, 1 month

[Seminaire-LRDE] Mercredi 11 décembre 2013 : Michael Wilkinson, Johann Bernoulli Institute, University of Groningen, The Netherlands

by Daniela Becker

Chers collègues, La deuxième session du séminaire Performance et Généricité du LRDE (Laboratoire de Recherche et Développement de l'EPITA) du mois de décembre aura lieu le Mercredi 11 décembre 2013 (14h--15h30), Salle L0 du LRDE. Au programme: * 14h-15h30: A ``Diplomatic'' Parallel Algorithm for the Component Trees of High Dynamic Range Images -- Michael Wilkinson --- Johann Bernoulli Institute, University of Groningen, The Netherlands http://www.cs.rug.nl/~michael/ Component trees are essential tools in several morphological processing methods, such as attribute filtering, or visualization, or the computation of topological watersheds. Algorithms for computation of these trees fall into two main categories: (i) Flood-filling algorithms, exemplified by the algorithms of Salembier et al (1998), Hesselink (2003), and Wilkinson (2011), and (ii) Union-find methods, such as Najman and Couprie (2006), and Berger et al (2007). As images become larger, and parallel computers become commonplace, there is an increased need for concurrent algorithms to compute these trees. The first such algorithm was published by Wilkinson et al in 2008, and was based on the divide-and-conquer principle. It splits up the data spatially, computes local component trees using any arbitrary algorithm which yields a union-find type representation of each node, and glues these together hierarchically. The main drawback of this method is that it does not work well on high-dynamic-range images, because the complexity of the glueing phase scales linearly with the number of grey levels. In the current work, carried out by Moschini, Meijster, and Wilkinson within the HyperGAMMA project, we develop a new algorithm for floating point or high-dynamic-range integer images. It works in a two-tier process, in which we first compute a pilot component tree at low dynamic range in parallel, with one grey level per processor using dynamic quantization, and Salembier's flood-filling method to build the local trees, and the previous parallellization scheme. After this, a refinement stage is performed. This refinement stage is based on the algorithm of Berger et al. As such, the new algorithm combines the two main types of algorithm in a single framework. Timings on images of up to 3.9 GPixel indicate a speed-up of up to 22 on 64 cores. The algorithm is more than 14x faster than the fastest sequential algorithm on the same machine. We will apply the new algorithm to astronomical data sets, including improvements to the SExtractor tool for object detection. The importance of the new algorithm extends beyond computation of component trees because it allows development of a fast alpha-tree algorithm suitable for any pixel-difference metric in case of vector images (i.e. not just $L_\infty$-based metrics on low dynamic range colour images). -- Michael Wilkinson obtained an MSc in astronomy from the Kapteyn Laboratory, University of Groningen in 1993, after which he worked on image analysis of intestinal bacteria at the Department of Medical Microbiology, University of Groningen, obtaining a PhD at the Institute of Mathematics and Computing Science, also in Groningen, in 1995. After this he worked as a researcher at the Johann Bernoulli Institute for Mathematics and Computer Science (JBI) both on computer simulations and on image analysis of diatoms. He is currently senior lecturer at the JBI, working on morphological image analysis and especially connected morphology. Apart from publishing in many journals and conferences, he edited the book ``Digital Image Analysis of Microbes'' (John Wiley, UK, 1998) with Frits Schut, and is member of the Steering Committee of ISMM. Pour plus de renseignements, consultez http://seminaire.lrde.epita.fr/. L'entrée du séminaire est libre. Merci de bien vouloir diffuser cette information le plus largement possible. -- Akim Demaille Akim.Demaille(a)lrde.epita.fr _______________________________________________ Seminaire mailing list Seminaire(a)lrde.epita.fr https://lists.lrde.epita.fr/listinfo/seminaire

11 years, 1 month

[Seminaire-LRDE] Mercredi 4 décembre 2013 : Juliusz Chroboczek, Laboratoire PPS, Université Paris-Diderot (Paris 7)

by Daniela Becker

Chers collègues, La prochaine session du séminaire Performance et Généricité du LRDE (Laboratoire de Recherche et Développement de l'EPITA) aura lieu le mercredi 4 décembre 2013 (11h--12h), Salle L0 du LRDE. Au programme : * 11-12h: CPC: Une implémentation efficace de la concurrence par passage de continuations -- Juliusz Chroboczek, Laboratoire PPS, Université Paris-Diderot (Paris 7) http://www.pps.univ-paris-diderot.fr/~jch/software/cpc/ CPC est une extension concurrente du langage C. Le code CPC, en style à threads, est traduit par le compilateur CPC en un code à style à événements; ce code peut ensuite être exécuté, au choix du programmeur, par des threads natifs « lourds » ou par un ordonnanceur à événements manipulant des structures de données extrêmement légères. Cette technique d'implémentation induit un style de programmation original, où les threads sont « gratuits ». Cependant, le programmeur peut choisir d'utiliser des threads natifs « lourds » lorsque c'est nécessaire, par exemple pour exploiter le parallélisme du matériel ou utiliser des bibliothèques bloquantes. La technique de compilation de CPC est basée sur des techniques formalisées et bien connues de la communauté de la programmation fonctionnelle, telles que la conversion en style à passage de continuations (CPS), le lambda-lifting, ou l'introduction de fonctions terminales. La correction de ces techniques a été prouvée formellement. Dans cet exposé, je donnerai quelques exemples du style typique de programmation en CPC tirées de Hekate, un seeder BitTorrent écrit en CPC. Je décrirai ensuite la transformation en style à passage de continuations et je décrirai la technique de traduction utilisée par le compilateur CPC. -- Juliusz Chroboczek est Maître de Conférences à l'Université Paris-Diderot (Paris 7). Il travaille sur les implémentations efficaces de la concurrence ainsi que sur la problématique du routage dans les réseaux à commutation de paquets. !! Attention : deuxième session en décembre avec un horaire différent (14h00 - 15h30) !! Le mercredi 11 décembre 2013, nous accueillons Michael Wilkinson, Johann Bernoulli Institute, University of Groningen, Pays-Bas, pour un exposé intitulé "A “Diplomatic” Parallel Algorithm for the Component Trees of High Dynamic Range Images". Pour plus de renseignements, consultez http://seminaire.lrde.epita.fr/. L'entrée du séminaire est libre. Merci de bien vouloir diffuser cette information le plus largement possible. -- Akim Demaille Akim.Demaille(a)lrde.epita.fr _______________________________________________ Seminaire mailing list Seminaire(a)lrde.epita.fr https://lists.lrde.epita.fr/listinfo/seminaire

11 years, 1 month

[Seminaire-LRDE] Prochaines sessions du séminaire Performance et Généricité

by Daniela Becker

Chers collègues, Petit rappel, demain, mercredi 13 novembre 2013 aura lieu l'exposé de Dodji Seketeli, Red Hat, intitulé "Address & Thread Sanitizer in GCC: Etat Actuel et Orientation Future" (11h--12h30), Salle Lα du LRDE. En décembre, nous organisons deux séminaires : * le mercredi 4 décembre 2013, 11h--12h, Salle L0 du LRDE, nous accueillons Juliusz Chroboczek, Laboratoire PPS, Université Paris-Diderot (Paris 7) pour nous parler de "CPC: Une implémentation efficace de la concurrence par passage de continuations" * le mercredi 11 décembre 2013 (horaire à définir), nous invitons Michael Wilkinson, Johann Bernoulli Institute, University of Groningen, Pays-Bas, pour nous faire un exposé intitulé "A “Diplomatic” Parallel Algorithm for the Component Trees of High Dynamic Range Images". Pour plus de renseignements, consultez http://seminaire.lrde.epita.fr/. L'entrée du séminaire est libre. Merci de bien vouloir diffuser cette information le plus largement possible. -- Akim Demaille Akim.Demaille(a)lrde.epita.fr _______________________________________________ Seminaire mailing list Seminaire(a)lrde.epita.fr https://lists.lrde.epita.fr/listinfo/seminaire

11 years, 1 month

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Annonce November 2013