DIGITAL LIBRARY
OPTIMIZING LEARNING OUTCOMES: COMBINING GAUSSIAN SPLATS, AND METAHUMAN AI IN MIXED REALITY ENVIRONMENTS
Munster Technological University (IRELAND)
About this paper:
Appears in: ICERI2024 Proceedings
Publication year: 2024
Pages: 3325-3333
ISBN: 978-84-09-63010-3
ISSN: 2340-1095
doi: 10.21125/iceri.2024.0860
Conference name: 17th annual International Conference of Education, Research and Innovation
Dates: 11-13 November, 2024
Location: Seville, Spain
Abstract:
The pursuit of optimizing educational outcomes has led to the exploration of various emerging technologies, with virtual reality at the forefront. This study investigates the synergistic potential of combining 3D photo-realistic digital landscapes, and metahuman AI to enhance learning experiences. Gaussian Splats offer sophisticated 3D visual representations of objects and landscapes, and metahuman AI introduces realistic interactive avatars. Together, these technologies can be used to create immersive educational 3D environments that not only engage learners but also have the potential to improve retention and comprehension. Our research focuses on the practical application of these technologies, aiming to transform modern education.

In the first stage of research, we have filmed and successfully rendered 3D photo-realistic digital landscapes using a process called Gaussian Splatting. We then imported the rendered 3D models into gaming engines and merged them together to create photo-realistic virtual production landscapes in a more efficient manner resulting in improved quality. This process not only enhances the visual fidelity of our projects but also lays a foundation for immersive content creation. In tandem, we have integrated virtual avatars i.e. metahumans with advanced body tracking using iPads, in doing this we have developed highly realistic and interactive avatars to integrate into these landscapes. By combining these technologies, we will outline the creation of bespoke virtual reality training environments, and our ongoing research methodology to optimize and enhance these workflows further.

We will be focusing our research on the bio-economy sector as a test case and how optimizing learning outcomes can be reached through virtual reality education. We aim to demonstrate how combining the technologies mentioned in this abstract can be used in multiple sectors and demographics to offer pedagogical advantages to immersive virtual reality applications. Throughout the course of our research, we will use a design sprint methodology to test and refine our workflow in developing virtual reality applications using the technologies mentioned in this abstract.
Keywords:
Technology, virtual reality, AI, education, gaussian splatting, 3D modeling.