Miguel Escrivá

Multiresolution 3d rendering on mobile devices

We present a client/server system that is able to display 3D scenes on handheld devices. At the server we extract the geometry that is visible for each client and send it. We also extract texture and material information. The clients, running on mobile devices, use that information to render realistic images. Our geometry extraction algorithm employs multiresolution and view-dependent simplification. We present results of our system running on PocketPC 2003 PDAs.

Adding new communication services to the FIPA message transport system

Agent communication is one of the most important aspects in the multi-agent system area. In recent years, several works have been developed that are related to the agent communication problem. This paper presents a new method for agent and agent platform communication in accordance with FIPA proposals. It uses the Jabber protocol as a new message transport protocol (MTP). This protocol provides additional services that are not included in the current standard FIPA MTP. It provides facilities for “presence notification”, “multi-user conference” and “security services“. As a result of this work, a new plug-in for the JADE platform that incorporates this transport protocol has been developed.

Modeling and Rendering of DPP-Based Light Fields

Autostereoscopic displays are a subject of recent research efforts in computer graphics. Such displays have to be fed graphics information in order to produce spatial images. This information is typically 4D radiance data called a light field. Traditionally light-field models were based on the two-plane parameterization. In this paper, however, we present a light-field representation that is based on the direction-and-point parameterization. This parameterization has certain uniformity properties that produce better rendering results. We describe the files and data structures needed to store the representation, and we introduce a rendering algorithm that takes advantage of the uniformity properties of the direction-and-point parameterization. Our algorithm runs in real time and renders light-field models that look like their geometric counterparts

Virtainer: graphical simulation of a container storage yard with dynamic portal rendering

The popularity of 3D graphics has grown exponentially in the latest years, which has lead to an increase in the range of applications where this kind of representations are used. The monitoring of industrial processes is an area where this type of simulation is rarely used, being much more common to show the information in traditional 2D interfaces. An application for data visualization in maritime container terminals is introduced here. We have developed a modular system adaptable to any stacked objects problem. This paper describes the architecture of our system, its features, and the graphics techniques applied to achieve a high frame rate and keep it independent of the data size.

Autostereoscopic Rendering of Multiple Light Fields

Light fields were introduced a decade ago as a new high‐dimensional graphics rendering model. However, they have not been thoroughly used because their applications are very specific and their storage requirements are too high. Recently, spatial imaging devices have been related to light fields. These devices allow several users to see three‐dimensional (3D) images without using glasses or other intrusive elements.

Continuous LODs and Adaptive Frame-Rate Control for Spherical Light Fields

Light fields are an image-based representation that represents objects using sets of digital images. Light fields are usually comprised of 4D radiance data that can be used for 3D rendering or 4D display on autostereoscopic and certain types of volumetric displays. We present a modeling and rendering system for spherical isotropic light fields. These are made of images captured with cameras placed on a spheres surface looking inwards. The system implements a multiresolution representation for both the spatial and the directional domains of the light field. This representation supports continuous levels of detail and adaptive frame rate control.

Un sistema de caché para la visualización interactiva de campos de luz de alta resolución

Los modelos de visualización basados en campos de luz almacenan valores de radiancia muestreados a lo largo de un conjunto de lineas orientadas en un espacio 4-dimensional. Esta elevada dimensionalidad junto al hecho de que estos modelos se suelen basar en un muestreo denso del campo de luz hace imposible la carga de un modelo completo directamente en memoria principal o memoria de la GPU para su visualización. Así pues, dicha información debe ser almacenada en memoria secundaria y se hace necesario un mecanismo adecuado para transferir esa gran cantidad de información desde la memoria secundaria hasta la memoria de la GPU a fin de visualizar estos modelos. En este trabajo presentamos un estudio teórico para la construcción de una caché orientada a la carga de la información de radiancia para campos de luz, junto a los resultados preliminares de su implementación. Esta caché se basa en una fragmentación de las imágenes que forman el conjunto de datos original para, utilizando un esquema de carga bajo demanda, conseguir maximizar el uso de los recursos disponibles y mejorar la velocidad de visualización de múltiples campos de luz. Esta aproximación permite además su combinación con técnicas de multirresolución que se suman para conseguir una mayor aceleración de la visualización.