Integrating multimedia documents in 3D city models for a better understanding of territories

TL;DR

This paper presents four methods to embed multimedia documents into 3D city models, enhancing urban understanding by providing contextual information and supporting spatial navigation-based document search.

Contribution

It introduces novel approaches for integrating multimedia data into 3D urban scenes and combines them with user guidance modes to improve territorial analysis.

Findings

Enhanced contextual understanding of urban areas

Effective integration of multimedia in 3D models demonstrated in Lyon

Supports spatial navigation for document retrieval

Abstract

Digital 3D representations of urban areas, through their growing availability, are a helpful tool to better understand a territory. However, they lack contextual information about, for example, the history or functionality of buildings. On another side, multimedia documents like images, videos or texts usually contain such information. Crossing these two types of data can therefore help in the analysis and understanding of the organization of our cities. This could also be used to develop document search based on spatial navigation, instead of the classical textual query. In this paper, we propose four approaches to integrate multimedia documents in a 3D urban scene, allowing to contextualize the scene with any type of media. We combine these integration approaches with user guidance modes that allows to guide the user through the consumption of these media and support its understanding of the territory. We demonstrate the usefulness of these techniques in the context of different projects within the Lyon area (France). The use of multimedia documents integrated into a digital tour allows, for example, the iconic buildings to be contextualised or to understand the evolution of a territory through time.