Research

Paper

AI LLM February 25, 2026

ViSTAR: Virtual Skill Training with Augmented Reality with 3D Avatars and LLM coaching agent

Authors

Chunggi Lee, Hayato Saiki, Tica Lin, Eiji Ikeda, Kenji Suzuki, Chen Zhu-Tian, Hanspeter Pfister

Abstract

We present ViSTAR, a Virtual Skill Training system in AR that supports self-guided basketball skill practice, with feedback on balance, posture, and timing. From a formative study with basketball players and coaches, the system addresses three challenges: understanding skills, identifying errors, and correcting mistakes. ViSTAR follows the Behavioral Skills Training (BST) framework-instruction, modeling, rehearsal, and feedback. It provides feedback through visual overlays, rhythm and timing cues, and an AI-powered coaching agent using 3D motion reconstruction. We generate verbal feedback by analyzing spatio-temporal joint data and mapping features to natural-language coaching cues via a Large Language Model (LLM). A key novelty is this feedback generation: motion features become concise coaching insights. In two studies (N=16), participants generally preferred our AI-generated feedback to coach feedback and reported that ViSTAR helped them notice posture and balance issues and refine movements beyond self-observation.

Metadata

arXiv ID: 2602.22077
Provider: ARXIV
Primary Category: cs.HC
Published: 2026-02-25
Fetched: 2026-02-26 05:00

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.22077v1</id>\n    <title>ViSTAR: Virtual Skill Training with Augmented Reality with 3D Avatars and LLM coaching agent</title>\n    <updated>2026-02-25T16:26:00Z</updated>\n    <link href='https://arxiv.org/abs/2602.22077v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.22077v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>We present ViSTAR, a Virtual Skill Training system in AR that supports self-guided basketball skill practice, with feedback on balance, posture, and timing. From a formative study with basketball players and coaches, the system addresses three challenges: understanding skills, identifying errors, and correcting mistakes. ViSTAR follows the Behavioral Skills Training (BST) framework-instruction, modeling, rehearsal, and feedback. It provides feedback through visual overlays, rhythm and timing cues, and an AI-powered coaching agent using 3D motion reconstruction. We generate verbal feedback by analyzing spatio-temporal joint data and mapping features to natural-language coaching cues via a Large Language Model (LLM). A key novelty is this feedback generation: motion features become concise coaching insights. In two studies (N=16), participants generally preferred our AI-generated feedback to coach feedback and reported that ViSTAR helped them notice posture and balance issues and refine movements beyond self-observation.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.HC'/>\n    <published>2026-02-25T16:26:00Z</published>\n    <arxiv:primary_category term='cs.HC'/>\n    <author>\n      <name>Chunggi Lee</name>\n    </author>\n    <author>\n      <name>Hayato Saiki</name>\n    </author>\n    <author>\n      <name>Tica Lin</name>\n    </author>\n    <author>\n      <name>Eiji Ikeda</name>\n    </author>\n    <author>\n      <name>Kenji Suzuki</name>\n    </author>\n    <author>\n      <name>Chen Zhu-Tian</name>\n    </author>\n    <author>\n      <name>Hanspeter Pfister</name>\n    </author>\n  </entry>"
}