ASTRAL
Augmented Self: TowaRds effective Avatars in augmented reaLity
Augmented Self: TowaRds effective Avatars in augmented reaLity
Abstract
Avatars can improve presence and performance in immersive environments and can alter users’ self-perception and behavior. Yet they remain less common in Augmented Reality (AR), where users can often see their physical body directly. ASTRAL investigates how to create compelling AR avatars and how those avatars affect users.
The project is organized around three complementary scientific axes:
- Technology — synthesizing and displaying avatars in both video see-through and optical see-through AR systems.
- Interaction — enabling users to act through an avatar while accounting for differences between the physical and virtual body and environment.
- Perception — studying embodiment, self-perception, self-confidence, performance, and the psychological and social effects of AR avatars.
Potential applications range from physical rehabilitation to sports-performance enhancement. The project combines Augmented Reality, computer vision, 3D interaction, avatar synthesis, embodied interaction, and perceptual evaluation.
People involved
Project coordinator
Lionel Dominjon, Clarté
Researchers at IMT Atlantique
PhD student
Juri Yoneyama’s doctoral work (October 2024–September 2027) studies perception and interaction with real and virtual hands and objects in AR, including aimed motions, grasping anticipatory behavior, and transfer effects between physical and augmented interaction.
Consortium
ASTRAL is a multidisciplinary consortium coordinated by Clarté, with IMT Atlantique / Lab-STICC, Centrale Nantes / LS2N PACCE, and Inria Rennes / Hybrid. The partners bring complementary expertise in AR, computer vision, 3D interaction, avatar synthesis, embodiment, and immersive-system evaluation.
Publications
Juri Yoneyama also contributed to A. Cheymol et al., “Measuring the Impact of Objects’ Physicalization, Avatar Appearance, and their Consistency on Pick-and-Place Performance in Augmented Reality,” IEEE Transactions on Visualization and Computer Graphics, 31(5), 2268–2276, 2025.
Funding
ASTRAL is funded by the French National Research Agency under reference ANR-23-CE33-0004. The project started in February 2024, runs for 42 months, and received an ANR grant of €613,639.


