Georgios Bardis

Intelligent Personalization in a Scene Modeling Environment

The current chapter presents an integrated approach towards the enrichment of a scene modeling information system with intelligent characteristics with respect to user preferences. The aim is to set a framework for this purpose covering the entire range of the scene modeling process, from the abstract definition to the final visual outcome. Necessary notions from preference modeling, decision analysis and machine learning are presented. Inherent difficulties of the current context are exposed and addressed. The architecture of an intelligent personalization module in the context of a scene modeling environment is defined and experimental results from its implementation and application are analyzed.

Collaborative Evaluation Using Multiple Clusters in a Declarative Design Environment

Collaborative Declarative Modeling combines the flexibility of abstract declarative description of a scene with the integration of views and opinions of multiple designers. One of the stages where this integration applies is the scene understanding phase, where solutions corresponding to the submitted description are visualized and evaluated. In the current work we propose a mechanism for collaborative solution evaluation based on previous feedback with respect to each collaborating designer’s preferences. Designers are grouped into clusters of preference revealing the proximity of their evaluations for previously seen solutions. These clusters are formed upon criteria ensuring wide inter-cluster preference representation and thorough intra-cluster compactness. Clusters subsequently participate through representative designers to the formation of the consensus solution set corresponding to the submitted description. The proposed mechanism allows adjustment of parameters controlling intuitive characteristics of the collaborative declarative design methodology such as importance of each designer and strictness in solution evaluation.

INTELLIGENT SOLUTION EVALUATION BASED ON ALTERNATIVE USER PROFILES

The MultiCAD platform is a system that accepts the declarative description of a scene (e.g. a building) as input and generates the geometric descriptions that comply with the specific description. Its goal is to facilitate the transition from the intuitive hierarchical decomposition of the scene to its concrete geometric representation. The aim of the present work is to provide the existing system with an intelligent module that will capture, store and apply user preferences in order to eventually automate the task of solution selection. A combination of two components based on decision support and artificial intelligence methodologies respectively are currently being implemented. A method is also proposed for the fair and efficient comparison of the results.

A Collaborative Framework for Virtual Globe Based 3D City Modeling

Urban planning and city modeling have gradually shifted, during the last decade, from highly demanding, in terms of required skills and supporting information, to tasks now supported by efficient, widely available applications, thus becoming popular and largely accessible. There are many GIS systems nowadays that offer a freely navigable three dimensional environment. This evolution of GIS systems has, in turn, led to the requirement for and creation of virtual 3D models of the ground and buildings. Online communities have created and distributed over the Internet libraries of georeferenced 3D models. The public is encouraged to participate in the design of 3D scenes and many companies offer free tools to facilitate the design of 3D models, specialized in buildings. In this paper, we present a collaborative approach for the construction of a city model, and its implementation through a prototype environment, employing freely available design tools. The prototype system comprises a collaborative database, supported by a web-based interface. Users are able to create and upload their models to the common database over the web, thus constructing a realistic 3D city model in a given area in a collaborative manner.

A GIS Platform for Automatic Navigation into Georeferenced Scenes Using GIS Scene Explorer (GIS-SE)

During the last decade the rapid evolution of GIS systems has led to enchanced accessibility and increased availability. The traditional content has been enriched by multimedia information and three dimensional models. Virtual copies of the real world have been created, widely known to the average user, yet the information retrieval and the efficient exploration of 3D large scale scenes are still pose serious problems. In this paper, a GIS platform for automatic navigation is presented, using a prototype application, entitled “GIS-Scene-Explorer” or GIS-SE, addressing these problems. Our platform is based on an existing GIS platform. Google Earth is used for this purpose. The prototype is a standalone application, adopting the Google Earth API to retrieve and visualize data. It takes advantage of both public remote database of the GIS system and a custom, collaborative database. Three dimensional models of buildings and multimedia data are stored to the collaborative database, operating under a server – client model architecture.

Abstract Description Refinement Using Incremental Learning and Scene Reconstruction

Declarative Modeling methodologies offer the designer the ability to describe a scene using abstract terms instead of precise geometric elements and properties. The price for this convenience is a large number of compliant geometric models, only a small subset of which is usually of practical interest for the designer. The task of solution evaluation can be tedious and time-consuming whereas the qualities that make these solutions stand out are not always straightforward. In the current work we outline the integration of a machine learning component, trained by user-approved solutions of previous descriptions, with a reconstruction component, able to discover relations and properties implied by the best solutions, into a unique module for description adaptation according to user preferences.

Administration and Exploitation of Qualitative Aspects in Declarative 3D Scene Synthesis

In this paper, we discuss the study and implementation of two alternative artificial intelligence methodologies in declarative scene synthesis. In order to achieve this, we employ two approaches towards the acquisition and exploitation of the desired qualitative aspects of the synthesis on behalf of the designer. Our work concerns the adaptation and application of a machine learning technique, as well as of an evolutionary search method, in the current context. The former refers to the gradual construction of a preference model, comprising an incrementally learning mechanism based on actual designer solution evaluation during regular system use. The latter approach models qualitative aspects through a multi-objective genetic algorithm variation based on the weighted sum method. The algorithm is applied during the generation and understanding phases of declarative modeling. This work provides evidence of performance improvement of declarative scene synthesis at the expense of additional designer feedback.

An Integrated Framework for Semantic Modelling and Decision Support in 3D Urban Virtual Environments

This work addresses the tasks of semantic modeling of an urban virtual environment and integration of such a model into a fully functional framework, offering visualization, navigation and querying of the corresponding content. The proposed semantic model adopts a series of qualitative and quantitative criteria for the acquisition and exploitation of buildings related knowledge. The model is subsequently encoded and integrated in an environment for accessing georeferenced 3D urban virtual content. This enhancement allows for improved search, routing and navigation capabilities, while it offers the potential of a semantic decision-support system with pragmatic knowledge-based functionality in a realistic operating environment.

An Intelligent GIS Database Framework Featuring Building Query Functionality Using n-grams Encoding and k-means Classification

In the current work we present the design and synthesis of a GIS framework with emphasis in building classification and identification. Its fundamental functionality comprises the submission of the geometric representation of a building as a query sample, subsequently encoded and employed by a classifying mechanism in order to discover similar occurrences within a pre-processed GIS-based building database. The encoding of both the query sample and the existing building database relies on n-grams whereas the classification and identification scheme combines k-means and a set of global and local features for the construction of similarity classes.

A formal framework for declarative scene description transformation into geometric constraints

The disambiguation of a declarative visual model is a crucial step towards the generation of its geometric equivalents. Any abstract description has to be ultimately translated into a concrete set of values or relevant constraints, solely relying on quantifiable model characteristics. To this day, there is no general consensus with respect to a unified, formally defined model for this disambiguation process or the desired quantified outcome. The current work sets the basis for a uniform transformation process and the corresponding formal constraint model, inclusive of a number of already existing approaches for declarative modelling. The applicability of the proposed framework is exhibited in an existing declarative modelling environment, explicitly demonstrating its implementation in the specific context.