Journal of Digital Landscape Architecture

Seite drucken

AI-Driven Avatars in Immersive 3D Environments for Education Workflow and Case Study of the Temple of Demeter, Greece

Generative Artificial Intelligence (AI) and Virtual Reality (VR) technologies are changing education and offer new opportunities for how people interact with environments. Technological advances over the past few decades have lowered barriers to creating virtual environments; however, there are still challenges, particularly when creating realistic virtual environments of real places. Realism is important as the fidelity of virtual environments influences user experience. In addition, methods and techniques that can facilitate ease of interacting with these environments are needed to streamline the user experience. One promising way to do this is incorporating AI-driven avatars into realistic scenes, allowing users to use natural language immersive learning experiences focused on sustainability education that incorporate realistic scenes to interact with and learn about the landscapes. To this end, we have developed workflows for design scenarios and natural interaction with AI avatars. This study created point cloud digital 3D models from photogrammetry and incorporated these into Unreal Engine 5. We then integrated generative AI avatars into the environment, enabling natural language interaction between users and an AI tutor. The integration facilitates interactive participation and enables highprecision digital reproduction of the physical environment. The novel proposed workflow is presented via a case study for a virtual study abroad experience in Naxos, Greece, using an AI-driven tutor to educate about the historical aspects of the island. This project provides the beneficial learning experience of study abroad experiences without the economic and environmental costs of sending students on field excursions. We recommend constructing immersive education experiences using real-world environments and natural AI-driven conversations and demonstrating its potential to revolutionize social interaction, historical heritage preservation, and sustainable pedagogy.

Autor / Author: Gao, Haoting; Lindquist, Mark; Serrano Vergel, Ramiro
Institution / Institution: University of Michigan, Ann Arbor/USA; University of Michigan, Ann Arbor/USA
Seitenzahl / Pages: 12
Sprache / Language: Englisch
Veröffentlichung / Publication: JoDLA – Journal of Digital Landscape Architecture, 9-2024
Tagung / Conference: Digital Landscape Architecture 2024 – New Trajectories in Computational Urban Landscapes and Ecology
Veranstaltungsort, -datum / Venue, Date: Vienna University of Technology, Austria 05-06-24 - 07-06-24
Schlüsselwörter (de):
Keywords (en): Generative artificial intelligence, AI-driven avatars, virtual reality, heritage preservation, immersive experience
Paper review type: Full Paper Review
DOI: doi:10.14627/537752059
8080 -