Nous nous intéressons au problème de la spécification du séquencement des actions dans un environnement virtuel collaboratif. Il s’agit de définir puis de contrôler ce qui peut ou doit se passer au cours de la simulation dans un contexte potentiellement multiutilisateur. Ceci passe, entre autres, par la spécification (puis l’exécution) d’un ensemble de scénarios possibles lors d’une session de simulation ainsi que par la distribution des actions réalisables entre les différents acteurs (réels ou virtuels) intervenant dans la simulation.Nous présentons #SEVEN, un modèle fondé sur les réseaux de Petri, permettant de décrire des agencements temporels et causaux des actions dans un environnement. #SEVEN est ensuite utilisé pour répondre aux problèmes de la spécification de l’ensemble des scénarios possibles et de la distribution des actions entre les acteurs. Les propriétés de #SEVEN en font un modèle capable de s’adapter facilement aux besoins de la spécification de scénarios notamment, car il permet de fournir différents niveaux de guidage. Il peut par exemple définir précisément les actions à réaliser et l’ordre qu’elles doivent avoir ou encore indiquer les changements d’état de l’environnement devant avoir lieu sans préciser les actions nécessaires. Ensuite, nous abordons le problème de la distribution des actions en proposant un modèle d’équipe permettant de modéliser les comportements et règles liés à l’organisation d’un groupe d’acteur. Ce modèle permet de faire évoluer les possibilités d’action offertes aux acteurs au cours de la simulation suivant leurs compétences, leur position dans l’équipe, les ressources auxquelles ils ont accès. / We are interested in the problem of the specification of the sequencing of actions in collaborative virtual environments. It is about defining, then controlling what must or can occur during the simulation in a potentially multi-user context. It is done, partly by the specification (and the execution) of a set of possible scenarios during a simulation session, and partly by the distribution of the feasible actions between the different actors (reals or virtuals) in the simulation.We present #SEVEN, a Petri nets based model, allowing to describe causal and temporal sequencing of actions in an environment. #SEVEN is then used to answer to the problem of the specification of the possible scenarios and of the distribution of the actions between the actors. #SEVEN’s properties made it a model able to be adapted to the needs of the specification of scenarios, especially because it can provide different guidance levels. As an example, it can define precisely the actions that need to be performed or indicates the changes that must occur in the state of the environment without defining precisely the actions to perform. Then, we address the problem of the distribution of the actions from perspective of the role theory, proposing a team model allowing to model the behaviours and rules related to a group of actors. This model allows to make evolve the possibilities of actions offered to the actors during the simulation depending on their abilities, their position in the team or the resources they have access to.

Dans des environnements de réalité virtuelle, les relations spatiales entre les objets transmettent les connaissances fondamentales sur l'environnement, par exemple la direction (« à gauche », « à droite »), la distance (« proche », « loin »), la topologie (« disjoint », « en contact »), et la projection (« entre », « entourée par »). La modélisation des relations spatiales est essentielle dans une grande variété d'applications de réalité virtuelle, tel que des environnements d'apprentissage humain, des musées virtuels, des systèmes d'aides à la navigation. Cependant, les relations spatiales ont été considérées comme des informations abstraites et donc, difficiles à spécifier. Afin d'aborder cette question, cette thèse propose une approche pour modéliser les relations spatiales entre les objets virtuels dans des environnements de réalité virtuelle. Nous formalisons un modèle formel des relations spatiales dédié aux environnements de réalité virtuelle. Ensuite, nous proposons un langage et un cadre pour spécifier les relations spatiales à un niveau conceptuel. Enfin, nous appliquons notre modèle pour spécifier les relations spatiales dans deux applications réelles : TPPhysique – un environnement d'apprentissage humain des exercices de physiques, et BrestCoz – une application pour visiter le port de Brest au 18ème siècle. Nous montrons que le langage proposé est une base pertinente pour spécifier des contraintes spatiales liées aux activités des agents et des utilisateurs dans les environnements de réalité virtuelle. / Within Virtual Reality Environments (VREs), spatial relationships among objects convey fundamental knowledge about the environment, namely direction ("left", "right", "front of"), distance ("near", "far"), topology ("inside", "disjoint"), and projection ("between", "surrounded by"). Modelling spatial relationships is critical in a variety of applications of VREs, such as human learning environments, virtual museums, or navigation-aids systems. However, spatial relationships have been considered as abstract information and thus, difficult to specify. Addressing this issue, this thesis proposes an approach to model spatial relationships among virtual objects in VREs. First, we formalise a formal model of spatial relationships dedicated to VREs. Second, we provide a language and a framework to specify spatial relationships at a conceptual level. Finally, we apply our model to specify spatial relations in two real applications: Virtual Physics Laboratory -- a VRE for learning physics, and BrestCoz -- an application for visiting Brest harbour in the 18th century. We claim that the proposed language is a relevant basis to specify spatial constraints related to activities of agents and users within VREs.

Virtual Reality (VR) technique has been studied and applied in a variety of academic and industrial fields. Because the advancement of the areas of Science, Technology, Engineering, and Mathematics (STEM) are important to national developments, literature on VR educational applications has focused almost exclusively on the areas of STEM. However, to date, no systematic investigation has considered the possibility of utilizing VR in painting appreciation. This researcher reconstructed a modern painting into a VR environment and employed constructivism with the model of art appreciation to investigate the effects of applying the VR technique in appreciating the painting on student cognitive and non-cognitive outcomes. Participants in the study included 60 undergraduates in the Department of Computer Graphics Technology at Purdue University. Quantitative analysis methods were used to analyze these students' responses to worksheets of painting appreciation. The findings from the research shows that under the situation without lectures/instructions: (a) the students using VR modern painting have significantly lower learning outcomes in interpreting the painting than the students using traditional 2D modern painting; (b) the students using VR modern painting have significantly higher levels of Interest/Enjoyment in motivation than the students using traditional 2D modern painting; (c) there is no significant effect of the VR modern painting on the intensity of induced emotions. This study discusses the implications of these findings and the influences of using the VR modern painting on art appreciation. Furthermore, this study not only offers the insights into the application of using VR technique in appreciating the modern painting, but also provides the recommendations for future teaching and/or learning in relevant environments or areas. Suggestions are also listed for future quantitative studies on VR application in painting appreciation.

The main purpose of this mixed methods research was to explore and analyze visitors’ overall experience while they attended a museum exhibition, and examine how this experience was affected by previously using a virtual 3dimensional representation of the museum itself. The research measured knowledge acquisition in a virtual museum, and compared this knowledge acquired between a virtual museum versus a real one, employing a series of questionnaires, unobtrusive observations, surveys, personal and group interviews related to the exhibition and the artist. A group of twenty-seven undergraduate students in their first semester at the College of Architecture and Design of the Autonomous University of the State of Mexico participated in the research, and were divided in two groups, one of which used a 3D virtual representation previous to the museum visit. Results show that participants who experienced the virtual museum concurred that using it was a positive experience that prepared them to go to the real museum because they knew already what they were going to find. Most of the participants who experienced the virtual museum exhibited an increased activity during their museum visit, either agreeing, being more participative, concurring and showing acceptance, asking questions, or even giving their opinion and analysis, disagreeing with the guide and showing passive rejection. Also participants from this group showed an increase on their correct answers to the knowledge acquisition questionnaires, going from 27% answers responded correctly in the pre-test, to 67% of correct answers after the virtual museum usage. The research attempted to show that experiencing a virtual museum can be similar to the experience in physical museum visits, not only engaging participants to go to the museum, but sometimes even offering a more functional way to deliver content. Results of this research evidence that using a virtual museum creates a positive impact in users before, during, and after the museum visit, and that it can be a good alternative, not only for educational, but for promotional and recreational and purposes.

Affordable and relatively accurate eye trackers are becoming more accessible for the mainstream user, consequently opening up new opportunities for utilizing eye gaze in interactive real-time applications. When extending gaze data from an eye tracker into real-time rendered 3D virtual environments, there are often techniques applied to correlate the gaze point to a 3D object. Techniques which can be referred to gaze-to-object mapping (GTOM) methods. One featured GTOM method, region-bound Interactors, provided by the eye tracking company Tobii AB, was extended in this thesis by utilizing a custom G-buffer in order to extract accurate shapes of individual objects in screen space. This adapted method was compared against two methods based on ray casting in a controlled 3D test environment consisting of three types of scenarios with varying conditions. Data was collected from a user study with 31 participants to assess if the Interactors method could be viable in 3D virtual environments. The extended method showed some promising result in relation to the other methods, but further work needs to be done to establish whether the method in question is practical for GTOM in 3D. / Relativt precisa eyetrackers med ett lågt pris blir allt mer tillgängliga för den vanlige användaren. Således öppnas nya möjligheter upp för att använda ögonspårning i interaktiva realtidsapplikationer. När man förlänger ögats blickpunkt in till virtuella realtidsrenderade 3D-miljöer, används ofta metoder för att kunna korrelera blicken till ett 3D-objekt. Dessa metoder kan benämnas som så kallade gaze-to-object mapping-metoder (GTOM). En GTOM-metod, regionsbundna Interactors, som tillhandahålls av eyetracking-företaget Tobii AB, kompletterades under detta examensarbete med en specifik G-buffer för att kunna extrahera exakta former av enskilda objekt på skärmen. Denna metod jämfördes med två metoder som var baserade på strålföljning i en kontrollerad testmiljö i 3D som bestod av tre typer av olika scenarier med varierande förhållanden. Information samlades in från en användarstudie med 31 deltagare för att kunna bedöma om Interactors-metoden är lämplig för GTOM i virtuella 3D-miljöer. Den utökade metoden uppvisade en del lovande resultat i förhållande till de andra två metoderna, men ytterligare forskning måste utföras för att kunna fastställa om huruvida den aktuella metoden är praktisk för GTOM i 3D.

As mobile devices still have limited battery life, processing power, memory, and display size, they cannot yet execute gaming applications with the same fidelity and quality as their PC counterparts. In response, researchers have recently performed research with the goal of the real-time delivery of game content specifically to fit within mobile devices’ limitations. In this thesis, we present a novel approach to tackling the streaming of objects to mobile devices. Our goal is to reduce the number of objects subject to streaming from the server to the target devices, while not violating the user-defined limitations through an efficient, context-aware 3D object selection and prioritization scheme. We take advantage of the game context to stream only the most relevant objects. Our evaluations have shown that this technique not only leads to better performance in general, but also increases the gameplay experience by helping the player to achieve a higher score.

Networked virtual reality environments including virtual worlds devoted to entertainment, online socializing and remote collaboration have grown in popularity with the rise of commercially available consumer graphics hardware and the growing ubiquity of the Internet. These virtual worlds are typified by a persistent simulated three-dimensional space that communicates over a computer network, where users interact with the environment and each other through digital avatars. Development of these virtual worlds challenges the limits of the networking infrastructure, 3D streaming graphics techniques, and the distributed computing design of the virtual world systems that manages the simulation. In this dissertation, we explore solutions to different aspects of the overall problem of developing a general purpose, networked virtual environment, focusing on the networking and software system issues. Specifically, we show how to improve the networking infrastructure to better support the high packet-rate traffic that is typical of virtual worlds, efficiently stream terrain data for remote rendering, and construct a dynamically adaptive distributed systems framework suitable for virtual world simulations.

Indiana University-Purdue University Indianapolis (IUPUI) / Depth estimation is one of the most important problems in computer vision. It has attracted a lot of attention because it has applications in many areas, such as robotics, VR and AR, self-driving cars etc. Using the defocus blur of a camera lens is one of the methods of depth estimation. In this thesis, we have researched this technique in virtual environments. Virtual datasets have been created for this purpose. In this research, we have applied graph cuts and convolutional neural network (DfD-net) to estimate depth from defocus blur using a natural (Middlebury) and a virtual (Maya) dataset. Graph Cuts showed similar performance for both natural and virtual datasets in terms of NMAE and NRMSE. However, with regard to SSIM, the performance of graph cuts is 4% better for Middlebury compared to Maya. We have trained the DfD-net using the natural and the virtual dataset and then combining both datasets. The network trained by the virtual dataset performed best for both datasets. The performance of graph-cuts and DfD-net have been compared. Graph-Cuts performance is 7% better than DfD-Net in terms of SSIM for Middlebury images. For Maya images, DfD-Net outperforms Graph-Cuts by 2%. With regard to NRMSE, Graph-Cuts and DfD-net shows similar performance for Maya images. For Middlebury images, Graph-cuts is 1.8% better. The algorithms show no difference in performance in terms of NMAE. The time DfD-net takes to generate depth maps compared to graph cuts is 500 times less for Maya images and 200 times less for Middlebury images.

No description available.

No description available.