Publications
Publications in Refereed Journals
We present a simplification algorithm for triangular meshes, implemented on the GPU. The algorithm performs edge collapses driven by a quadric error metric. It uses data parallelism as provided by OpenCL and has no sequential segments in its main iterative structure in order to fully exploit the processing power of the GPU. Our implementation produces results faster than a corresponding sequential implementation and the resulting models are of comparable quality.
Air traffic has increased substantially since the introduction of deregulation in both the USA and the European Union. Moreover, aircraft accidents involving fatalities have exhibited a downward trend over time. Still, a series of recently publicized accidents has raised again a serious issue, namely whether cost reduction in a deregulated aviation environment is achieved at the expense of safety standards.
To address this question, the paper proposes a mathematical model, which highlights the relationship between competitive behaviour and tort liability. The model has important policy implications suggesting that the level of airline penalisation should be reduced when the market conduct is restricted and conversely.
Progressive meshes are an established tool for triangle mesh simplification. By suitably adapting the simplification process, progressive hulls can be generated which enclose the original mesh in gradually simpler, nested meshes.
We couple progressive hulls with a selective refinement framework and use them in applications involving intersection queries on the mesh. We demonstrate that selectively refinable progressive hulls considerably speed up intersection queries by efficiently locating intersection points on the mesh. Concerning the progressive hull construction, we propose a new formula for assigning edge collapse priorities that significantly accelerates the simplification process, and enhance the existing algorithm with several conditions aimed at producing higher quality hulls. Using progressive hulls has the added advantage that they can be used instead of the enclosed object when a lower resolution of display can be tolerated, thus speeding up the rendering process.
We present an algorithm for ray-tetrahedron intersection. The algorithm uses Plücker coordinates to represent the ray and the edges of the tetrahedron and employs a robust and efficient test to determine the intersection. The algorithm is highly optimized and provides a significant performance increase over related algorithms.
This paper presents the results of research work in the field of heterogeneous database integration. The CORBA architecture is used for constructing generalized interfaces for accessing heterogeneous data sources, for combining them, and for the transparent user access to them.
This is an expanded version of paper [C1], selected for publication in the Journal of the Society for Design and Process Science.
This paper contains the results of research work in the field of combinatorial designs. We study two-variable orthogonal designs of order 36, and by combining their theoretical properties with exhaustive search we declare conditions for their existence.
Publications in Refereed Conference Proceedings
In this paper we examine the suitability of the Google Cardboard as a means for the delivery of personalized cultural experiences. Specifically, we develop the content and create the application required in order to provide highly personalized visits to the Archaeological Museum in Tripolis, Greece. We also examine the usability issues related to the use of Google Cardboards. Early results are promising, and based on them we also outline the next steps ahead.
Tag clouds provide an excellent means of visualization of weighted semantic information. With their generation depending on given or calculated weights, their use is not possible when these weights are not known with certainty. In this paper we propose an extension of tag clouds to support the notion of uncertainty and explore some properties of this new representation. Furthermore, we present a tool implementing it.
Tag clouds provide an excellent means of visualization of weighted semantic information. When, on the other hand, this information is not definitive but is rather accompanied by a measurable degree of uncertainty, conventional tag clouds are no longer suitable visualization tools. In this paper we extend the conventional approach to tag cloud generation and propose the utilization of the degree of opaqueness as a means to visualize the degree of certainty. In order to experimentally assess the efficacy of the proposed approach we have developed the corresponding software tools and have applied the conventional and proposed approached to tag visualization in a real life scenario of probabilistic data.
The advent of mobile technologies has created opportunities for more effective delivery of cultural heritage information to all interested parties, and especially tourists who are on-the-move. Many people nowadays are equipped with smartphones or tablets equipped with location tracking devices such as GPS and compasses and having internet access available; these smartphones can be used for efficient delivery of cultural heritage information, in a personalized and timely fashion. Moreover, the computing resources of these smartphones are adequate to enable the use of augmented reality techniques, which provide a lively and engaging experience to the users. However, the same content still needs to be disseminated to users of stationary computers. In this paper, we report on the design and development of a mobile-enabled platform for presenting and disseminating cultural heritage information enhanced with augmented reality and integration with social networks.
Simulators are an inseparable part of the design and evaluation of distributed multi-agent protocols. In this work we put forward PViz, a novel visualisation tool built for one of the most prominent, publicly available P2P simulators, PeerSim. Our tool provides network visualisation for different overlays, interaction with the simulation through scenario re-playability and stepwise execution, and intuitive visualisation features such as panning and zooming in/out of the network, color-coding of nodes, and event alerting.
In this paper, we present the vision of an open source learning analytics platform, able to harvest data from different sources, including e-learning platforms and environments, registrar's information systems, alumni systems, etc., so as to provide all stakeholders with the necessary functionality to make decisions on the learning process. The platform's architecture is modular, allowing the introduction of new functionality or connection to new systems to collect needed data. All data can be analyzed and presented though interactive visualizations to find correlations between metrics, to make predictions for students or student groups, to identify best practices for instructors and let them explore 'what-if' scenarios, to offer students personalized recommendations and personalized detailed feedback, etc. Our objective is to inform and empower all stakeholders to improve the learning experience.
Vector fields produced by experiments or simulations are usually extremely dense, which makes their ma-nipulation and visualization cumbersome. Often, such fields can be simplified without much loss of infor-mation.
A simplification method for 3D vector fields defined over tetrahedral meshes is presented. The underlying tetrahedral mesh is progressively simplified by successive half-edge collapses. The order of col-lapses is determined by a compound metric which takes into account the field and domain error incurred as well as the quality of the resulting mesh. Special attention is given to the preservation of the mesh boundary and of critical points on the vector field. A tool has been developed for the measurement of the difference between two vector fields over tetrahedral meshes, and it is used to quantify the simplification error.
This paper presents the results of research work in the field of heterogeneous database integration. The CORBA architecture is used for constructing generalized interfaces for accessing heterogeneous data sources, for combining them, and for the transparent user access to them.
Books and Lecture Notes
This textbook was developed within the “Kallipos” initiative. It aims to present the theoretical background, the technologies, the design, implementation and evaluation techniques as well as applications of Virtual Reality systems. This textbook is addressed to both undergraduate and postgraduate students who wish to become familiar with the development of Virtual Reality systems. It covers all the related topics of Matematics, Electrical Engineering, Ergonomics, Psychology and Computer Science that are necessary to understand the specificities of developing a Virtual Reality system.
This book is a comprehensive visual computing textbook, dealing with the modeling and synthesis of visual data by means of computers. The book is aimed at undergraduate and graduate students taking computer graphics and visualization courses. The book concentrates on established principles and algorithms as well as novel methods that are likely to leave a lasting mark on the subject.
The book has been adopted as the textbook of several Graphics and Visualization courses in universities worldwide.
Contents: General elements of programming languages — Structured programming (basic commands, expressions, control and iteration structures, arrays) — Pseudocode, flow diagrams — Procedures, functions, recursion — Searching and sorting algorithms.
Until the publication of [B5], this was the most complete and current Greek book on Computer Graphics. The book covers a broad material, from basic mathematical principles, geometric transformations and two-dimensional algorithms, to advanced lighting algorithms and animation. It was used as the main textbook in most Graphics courses offered by universities in Greece.
Chapter contents: Bézier curves — B-Spline curves — Interpolation curves — Bézier, B-Spline, Interpolation surfaces.
Book contents: Vector spaces — Matrices — Complex numbers — Basic elements of Topology — Real functions of one and multiple variables — Optimisation — Differential equations — Applications in Economics.
Miscellaneous publications
GraphViz is a set of programs for graph visualization. Its capabilities are available to C++ through the Boost Graph Library (BGL). In this article we present an introduction to the use of GraphViz as well as examples of graphs constructed programmatically, in C++ and BGL, and visualized using GraphViz. The examples employ advanced object oriented programming techniques and utilize other parts of the Boost library.
Theses
This thesis studies the simplification of triangular and tetrahedral meshes by the use of techniques based on successive edge collapses, as well as the exploitation of the generated multiple levels of detail (progressive meshes) for the effective processing of the models.
Regarding triangular meshes, a method for the construction of progressive hulls by suitable edge collapses is presented. The generated hulls are used for the acceleration of intersection tests between the initial mesh and a line.
Regarding tetrahedral meshes, meshes with associated vector fields are simplified. Progressive tetrahedral meshes are constructed by taking into account, while collapsing edges, both the geometry of the mesh and the associated field.
Finally, an efficient algorithm for computing ray-tetrahedron intersection is presented, which exploits Plücker coordinates to accelerate computations; this algorithm may be used for the efficient processing of progressive tetrahedral meshes.
Τhis thesis studies Bézier and B-Spline parametric curves and surfaces, their properties and methods for their computation. Furthermore, algorithms for their display are developed, using object-oriented techniques integrated with the GLOOP library used at the Department of Computer Science, University College London, for teaching Computer Graphics courses.