Next Generation XR Systems—Large Language Models Meet Augmented and Virtual Reality
Authors:
- Muhamamd Zeshan AfzalMAMuhamamd Zeshan Afzal
- Sk Aziz AliSASk Aziz Ali
- Didier StrickerDidier Stricker
- Peter EisertPEPeter Eisert
- Anna HilsmannAHAnna Hilsmann
- Daniel Perez-MarcosDPDaniel Perez-Marcos
- Marco BianchiMBMarco Bianchi
- Sonia Crottaz-HerbetteSonia Crottaz-Herbette
- Roberto De IorisRIRoberto De Ioris
- Eleni ManginaEleni Mangina
- Mirco SanguinetiMSMirco Sanguineti
- Ander SalaberriaASAnder Salaberria
Top Cited Papers
Open Access
- 6 March 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Graphics and Applications
- Vol. 45 (1) , 43-55
Abstract
Extended Reality (XR) is evolving rapidly, offering new paradigms for humancomputer interaction. This position paper argues that integrating Large Language Models (LLMs) with XR systems represents a fundamental shift toward more intelligent, context-aware, and adaptive mixed-reality experiences. We propose a structured framework built on three key pillars: (1) Perception and Situational Awareness, (2) Knowledge Modeling and Reasoning, and (3) Visualization and Interaction. We believe leveraging LLMs within XR environments enables enhanced situational awareness, real-time knowledge retrieval, and dynamic user interaction, surpassing traditional XR capabilities. We highlight the potential of this integration in neurorehabilitation, safety training, and architectural design while underscoring ethical considerations such as privacy, transparency, and inclusivity. This vision aims to spark discussion and drive research toward more intelligent, human-centric XR systems.Keywords
Funding Information
- European Commission (101135724)
This publication has 19 references indexed in Scilit:
- A survey on multimodal large language modelsNational Science Review, 2024
- Groundhog Grounding Large Language Models to Holistic Segmentation2024
- Animatable Virtual Humans: Learning Pose-Dependent Human Representations in UV Space for Interactive Performance SynthesisIEEE Transactions on Visualization and Computer Graphics, 2024
- Unsupervised learning of style-aware facial animation from real acting performancesGraphical Models, 2023
- I2MVFormer: Large Language Model Generated Multi-View Document Supervision for Zero-Shot Image Classification2023
- The use of virtual reality in the rehabilitation of aphasia: a systematic reviewDisability and Rehabilitation, 2022
- Neural Head Avatars from Monocular RGB Videos2022
- Capture and 3D Video Processing of Volumetric Video2019
- Surface Motion Capture Animation SynthesisIEEE Transactions on Visualization and Computer Graphics, 2018
- The Uncanny Valley [From the Field]IEEE Robotics & Automation Magazine, 2012