Title: Real-time video avatar generation method for realistic communication in the metaverse

Authors: Momoka Hagihara; Naho Kuriya; Tomoyuki Ishida

Addresses: Graduate School of Engineering, Fukuoka Institute of Technology, Fukuoka, Japan ' Graduate School of Engineering, Fukuoka Institute of Technology, Fukuoka, Japan ' Graduate School of Engineering, Fukuoka Institute of Technology, Fukuoka, Japan

Abstract: In this study, we implemented a real-time video avatar generation method using image processing and machine learning to develop a new advanced communication tool that enables information sharing and decision-making. Using the proposed method, users can easily send their real-time video avatars to the metaverse. We evaluated a questionnaire survey on 12 subjects to evaluate the effectiveness of the proposed real-time video avatar generation method. This evaluation was conducted based on the sense of being in the same room, degree of concentration, degree of communication of non-verbal information, ease of timing utterances and naturalness of conversation for 12 subjects. The 12 subjects held discussions using video conferencing and our proposed metaverse conferencing systems with real-time video avatars and answered some evaluation questionnaires. The evaluation results revealed that our proposed metaverse conferencing system using real-time video avatars is superior to the video conferencing system in all evaluation items.

Keywords: metaverse; real-time video avatar; virtual reality; image processing; space sharing.

DOI: 10.1504/IJGUC.2025.145174

International Journal of Grid and Utility Computing, 2025 Vol.16 No.2, pp.105 - 116

Received: 31 Oct 2023
Accepted: 31 Jan 2024
Published online: 24 Mar 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article