Immersive Visualization of Virtual 3D City Models and its Applications in E-Planning

Introduction

“Major urban development projects extend over prolonged timescales, involve a large number of stakeholders, and necessitate complex decision making” (Isaacs et al., 2011). The most important decisions frequently have to be taken at an early stage of the project (Hunt et al., 2008) taking into account topological and geometrical characteristics as well as economic, social, and cultural factors (Hamilton et al., 2005). At final stages, projects typically involve external stakeholders such as the general public requiring a convincing, close-to-reality presentation of plans, variants, and processes. In both cases, fast access to urban spatial information and effective communication tools are needed. However, today’s geographic information systems (GIS) are often dominated by the view of experts and are still technically based on 2D concepts, while the underlying data has three or more dimensions (Isaacs et al., 2011).

Virtual 3D city models are essential for effective communication of complex three-dimensional urban spatial information. An increasing number of applications and systems use virtual 3D city models to integrate, manage, and visualize complex 2D and 3D urban geodata as well as associated geo-referenced thematic data (e.g., Autodesk Infrastructure Modeler, CityGRID, and CityServer3D). A growing number of cities are creating and continuing virtual 3D city models as a fundamental 3D geodata resource (Döllner et al., 2006). Meanwhile, the Open Geospatial Consortium has established the international encoding standard CityGML (Kolbe, 2009) for the representation, storage, and exchange of virtual 3D city and landscape models, implemented as an application schema of the Geography Markup Language (GML).

Virtual 3D city models are used in various application fields, e.g., driving simulations (Randt et al., 2007), disaster management (Lapierre & Cote, 2007), and e-planning (Knapp & Coors, 2008; Weber et al., 2009). Knapp and Coors (2008) developed an application for public participation via a web-browser that visualizes a virtual 3D city model to ease the understanding of planning proposals. Weber et al. (2009) use virtual 3D city models to simulate city development. Ball et al. (2007) proposed a 3D visualization system for landscape planning. While all these applications, as well as commercial tools like the Infrastructure Modeler, could be used with high resolution displays and projectors, they cannot be directly used for fully immersive virtual environments like CAVEs due to both technical and conceptual restrictions. In particular, they lack advanced visualization and rendering techniques, such as satisfactory real-time photorealistic rendering, assistive interaction techniques, and an immersion-supporting soundscape, and thus in many cases do not meet the expectations of stakeholders. Isaacs et al. (2011) present a decision support tool that supports power walls and stereoscopic rendering. Although some advanced rendering techniques (e.g., water rendering) are implemented it still lacks other crucial techniques, e.g., ambient occlusion, dynamic sky rendering, and a 3D soundscape.

Virtual environments (VEs) represent artificial environments that are based on interactively visualizing virtual 3D models; they have manifold fields of application in Virtual Reality (VR), Augmented Reality (AR), and Mixed Reality (MR). As Brunnett et al. (2008) point out, “an important aspect of VR-based systems is the stimulation of the human senses – typically sight, sound, and touch – such that a user feels a sense of presence (or immersion) in the virtual environment.” For applications in e-planning, VEs require a focus on that sense-of-presence, in particular, to support not only experts but also non-experts to examine spatial structure, appearance, and relationships of plan models “in situ.” Furthermore, e-planning typically includes collaborative processes that involve a large number of heterogeneous stakeholders, whereby decisions have to be understood, discussed, and made by non-experts as well as by experts (Ross et al., 2007).