EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Generating interaction gestures in dyadic conversations using a diffusion model

Yuya Okadome iD, Yazan Alkatshah iD, Yutaka Nakamura

PLOS One · 2025

DOI: 10.1371/journal.pone.0339579 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, 2025. Verfasst von Yuya Okadome und 2 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Engineering, Tokyo University of Science, Katsushika, Tokyo, Japan. Hauptthemen laut MEDLINE: Computer Graphics, Communication, Avatar, Generative Artificial Intelligence. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Japan Science and Technology Corporation, Japan Society for the Promotion of Science.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

As expectations for computer graphic (CG) avatars and conversational robots increase, enhancing dialogue skills via multimodal channels is crucial for achieving fluent interactions with humans. Thus, automatic interaction motion generation is essential for autonomous conversation systems. Natural motion generation, such as appropriate nodding, requires considering the behavior and voice of the conversation partner. However, current models generate motion from audio or text, neglecting interaction factors. In this study, we implemented an interaction diffusion model (IDM) that uses a diffusion approach and masking features to generate interaction behaviors for dyadic conversation. IDM accounts for two participants, using masks to generate features from conditional inputs. This allows for accommodating conditions like missing features and forecasting without retraining. The experimental results suggests that the model generates the human-like behaviors during conversation in 30 ms.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Computer GraphicsCommunicationAvatarGenerative Artificial IntelligenceFeasibility StudiesHumansMotionVideo RecordingMotion CaptureModels, Theoretical

Autor:innen und Institutionen

  • Faculty of Engineering, Tokyo University of Science, Katsushika, Tokyo, Japan.
  • RIKEN Information R&D and Strategy Headquarters, Sorakugun, Kyoto, Japan.
  • Hiroshi Ishiguro Laboratories, Advanced Telecommunications Research Institute International, Sorakugun, Kyoto, Japan.

ORCID-Profile: Yuya Okadome, Yazan Alkatshah

Förderung

  • Japan Science and Technology Corporation
  • Japan Society for the Promotion of Science

Zitieren

APA

Okadome, Y., Alkatshah, Y., & Nakamura, Y. (2025). Generating interaction gestures in dyadic conversations using a diffusion model. PLOS One. https://doi.org/10.1371/journal.pone.0339579

Vancouver

Okadome Y, Alkatshah Y, Nakamura Y. Generating interaction gestures in dyadic conversations using a diffusion model. PLOS One. 2025. doi:10.1371/journal.pone.0339579

BibTeX
@article{Okadome2025,
  author = {Okadome, Yuya and Alkatshah, Yazan and Nakamura, Yutaka},
  title = {Generating interaction gestures in dyadic conversations using a diffusion model},
  journal = {PLOS One},
  year = {2025},
  doi = {10.1371/journal.pone.0339579}
}
Kostenlos starten →