New dataset and generative pipeline enhance human-to-robot handover prediction

Researchers have developed Hand2Bot, a new dataset and generative pipeline called PassGen, aimed at improving human-to-robot object handover prediction. PassGen utilizes Stable Video Diffusion and an Intention-Aware Temporal Face Encoder to create realistic handover sequences, addressing the scarcity of data and the sim-to-real gap in human-robot collaboration. The system incorporates a morphology-based depth editing strategy to mimic real-world sensor noise, enabling more robust zero-shot transfer and earlier intention anticipation for socially aware robotic behavior. AI

IMPACT Enhances human-robot collaboration by improving intention anticipation and enabling socially aware robotic behavior in shared workspaces.

RANK_REASON The cluster describes a new dataset and generative pipeline for a specific research problem in robotics, published on arXiv. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• Hand2Bot
• Intention-Aware Temporal Face Encoder
• Stable Video Diffusion

• paper
• product
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →