Generative Tutorial: Towards Live Contextualized Visual Instructions for Physical Tasks
Researchers prototype an AR system that generates live, workspace-specific visual instructions for physical tasks.
The paper introduces Generative Tutorial, a framework for live visual instruction that depicts intended actions inside the user's own workspace. A formative review of image and video generation across 15 physical tasks informed an augmented-reality prototype that uses observed context and predicted outcomes. In a 24-participant lab study, the system improved task quality, perceived workspace correspondence, and step-confirmation time versus pre-authored guidance.
- Framework generates goal images and demos inside the user's own workspace
- Formative tests covered image and video generation across 15 physical tasks
- A 24-person study beat pre-authored guidance on quality and confirmation time
- Generation errors and visual resemblance shaped trust and interpretation
From the original article
1 min read at arxiv.orgVisual instructions for physical tasks are typically authored in one context and followed in another, requiring users to translate demonstrated tools, materials, and spatial relationships into their own environment. We introduce Generative Tutorial, a conceptual framework for live visual instruction that depicts intended outcomes and actions within the user's environment and task flow.
By Muzhe Wu · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.