Centre de documentation
scientifique

Real-time depth-image-based rendering for viewpoint-variable display on mobile devices / Shuoran Yang (2016)  

Public Titre : Real-time depth-image-based rendering for viewpoint-variable display on mobile devices Type de document : Mémoire Auteurs : Shuoran Yang, Auteur Editeur : Zurich : Eidgenossische Technische Hochschule ETH - Ecole Polytechnique Fédérale de Zurich EPFZ Année de publication : 2016 Importance : 36 p. Format : 21 x 30 cm Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Termes IGN] appareil portable [Termes IGN] petit écran [Termes IGN] réalité augmentée [Termes IGN] rendu réaliste [Termes IGN] temps réel [Termes IGN] visualisation de données [Vedettes matières IGN] Géovisualisation Index. décimale : 37.60 Géovisualisation Résumé : (auteur) The ability to adjust the display according to the user viewpoint gives a more immersive user experience to augmented reality applications. In this thesis, we propose a pipeline for rendering images to variable viewpoints from a video captured on a mobile device with an arbitrarily aligned camera sensor. We present algorithms to handle depth map restoration, disocclusion and synchronization issues to minimize artifacts in the rendering of virtual images. This pipeline is implemented in the Project Tango framework where an RGB-IR pixel sensor is used to capture depth map and color image inputs. At the end, we demonstrate the results with a video recorded on the Project Tango tablet. The results show that the pipeline is capable of rendering realistic images that have the potential of being used for augmented reality applications. Numéro de notice : 17204 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Mémoire masters divers En ligne : http://dx.doi.org/10.3929/ethz-a-010634330 Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=81185

Mapping literature : spatial data modelling and automated cartographic visualisation of fictional spaces / Anne-Kathrin Weber (2014)  

Public Titre : Mapping literature : spatial data modelling and automated cartographic visualisation of fictional spaces Type de document : Thèse/HDR Auteurs : Anne-Kathrin Weber, Auteur Editeur : Zurich : Eidgenossische Technische Hochschule ETH - Ecole Polytechnique Fédérale de Zurich EPFZ Année de publication : 2014 Collection : Dissertationen ETH num. 21530 Note générale : bibliographie A dissertation submitted to ETH Zurich for the degree of doctor of sciences Langues : Anglais (eng) Descripteur : [Termes IGN] analyse spatiale [Termes IGN] géovisualisation [Termes IGN] incertitude géométrique [Termes IGN] littérature [Termes IGN] modèle conceptuel de données localisées [Termes IGN] représentation cartographique [Vedettes matières IGN] Géovisualisation Index. décimale : 37.60 Géovisualisation Résumé : (auteur) Mapping literature addresses the issue of spatial analysis and visualisation of fictional, narrated spaces. Compared to the geography and phenomena of the ‘real world’, the ‘fictional world’ is distinguished by a fragmentary existence with vague boundaries and spaces that are often hard to localise, if at all. Furthermore, not only such spaces highly vary in size and detail, but they can also be transformed and remodelled by the author and can be linked to any time period. So far, literary scholars have tried to create maps with their own limited resources to analyse and explain where fiction is set and why there. However, a concise theory and convincing instruments to acquire and visualise this specific kind of geography is still lacking. This thesis devotes to the issue of how to structure and visualise literary spaces with the help of cartographic means. This work originates from the interdisciplinary project “A literary Atlas of Europe” initiated by and in close cooperation with experts of the literary studies researching the geography of literature. The main objectives are the development of a generally applicable data model for spaces of narrative, fictional literature such as novels, novellas or legends, including a system where data can be collected and geocoded. Furthermore, this work strives for adequate cartographic visualisation of uncertain, fragmentary literary spaces, which visually meet the requirements of the aforementioned inherent rules. The resulting spatial data model distinguishes four main categories of fictional spaces. The classification is done from the view point of the book’s characters: first, places where characters are present and acting (settings); second, places characters long for, remember or dream of (projected places); third, places they move through (routes); and finally places that are only mentioned by the author, having a marginal meaning for the character (marker). Visualisation methods for all four categories in consideration of several specific properties are suggested and implemented into a specially developed web map service to allow automatic map generation. In addition to information that can be interactively requested in the map application, symbols were introduced to visually differentiate between subcategories of projected places without having to forgo current map information. Further efforts were made to enhance the automated map visualisation by applying spatial deformation to better utilise the map space. This was realised through density calculation combined with cartogram algorithms. This cumulated thesis is based on four scientific publications and framed by an introductory and a concluding section. The included articles are structured in three chapters. Chapter 3 covers the data model and acquisition, Chapter 4 addresses the data characterisation and uncertainty visualisation and finally in Chapter 5 two articles focus on extending visualisation approaches. With the data acquisition and visualisation tools developed within this dissertation project, literary scholars focussing on literary geography can now create automatically visualised literary maps of improved quality. By formalising spatial concepts of literary geography within a data model, the scholar’s requirements are transferred into digital tools. It is hoped that such tools promote the usage of literary maps by facilitating their creation in digital form. In any case, the formalisation contributes to the development of an overarching theory for an advanced cartographic approach in literary geography. This results not only in possibilities to look at literature from another point of view, but moreover it promotes to ask new questions and facilitates to find an answer for them. Finally, this cartographic tool reveals invisible cultural layers of landscapes. In addition to the visible reality, it is the invisible cultural influence that has a lasting impact to a region. Such influences are increasingly encouraged by the UNESCO and defined by the term “associative landscapes. Numéro de notice : 17205 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Thèse étrangère Note de thèse : doctoral dissertation : Sciences : ETH Zurich : 2014 En ligne : http://dx.doi.org/10.3929/ethz-a-010106067 Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=81186

Data visualisation / Nathan Yau (2013)

Public Titre : Data visualisation : de l'extraction des données à leur représentation graphique Type de document : Monographie Auteurs : Nathan Yau, Auteur Editeur : Paris : Eyrolles Année de publication : 2013 Importance : 355 p. Format : 19 x 23 cm ISBN/ISSN/EAN : 978-2-212-13599-2 Langues : Français (fre) Descripteur : [Termes IGN] collecte de données [Termes IGN] extraction de données [Termes IGN] modèle logique de données [Termes IGN] représentation graphique [Termes IGN] visualisation de données [Vedettes matières IGN] Géovisualisation Index. décimale : 37.60 Géovisualisation Résumé : (Editeur) L'essor des nouvelles technologies et du Web a extraordinairement accéléré et simplifié la collecte, le stockage et l'accès aux données. Analysées et présentées de façon explicite et sensée, elles contribuent à faciliter la prise de décision, partager les connaissances et les idées, porter à un regard plus objectif sur le monde. La data visualisation est ainsi devenue une discipline à part entière, outil privilégié des datajournalistes, scientifiques, statisticiens, ingénieurs, graphistes, designers, chercheurs en sciences de l'information, spécialistes du marketing. Pionnier de cette approche innovante, Nathan Yau présente dans cet ouvrage les meilleurs moyens de collecter, d'explorer, d'analyser et de représenter de façon créative de larges ensembles de données. Note de contenu : Introduction 1. Raconter une histoire avec les données 2. Manipulation de données 3. Choix des outils pour la visualisation des données 4. Visualisation des modèles temporels 5. Visualisation des proportions 6. Visualisation des relations 7. Identification des différences 8. Visualisation des relations spatiales 9. Concevoir avec un objectif Numéro de notice : 22392 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Monographie Accessibilité hors numérique : Accessible à Georges Pérec (Id UGE) Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=78997