IMPACT-HOI: Supervisory Control for Onset-Anchored Partial HOI Event Construction
This work addresses the need for efficient, high-quality annotation of human-object interactions in egocentric video for robot learning, offering a supervisory control approach that balances automation and human input.
IMPACT-HOI introduces a mixed-initiative framework for annotating egocentric procedural video with structured event graphs for Human-Object Interactions, reducing manual annotation actions by 13.5% and achieving a 46.67% event match rate with zero confirmed-field violations.
We present IMPACT-HOI, a mixed-initiative framework for annotating egocentric procedural video by constructing structured event graphs for Human-Object Interactions (HOI), motivated by the need for high-quality structured supervision for learning robot manipulation from human demonstration. IMPACT-HOI frames this task as the incremental resolution of a partially specified, onset-anchored event state. A trust-calibrated controller selects among direct queries, human-confirmed suggestions, and conservative completions based on empirical annotator behavior and evidence quality. A risk-bounded execution protocol, utilizing atomic rollback, ensures that human-confirmed decisions are preserved against conflicting automated updates. A user study with 9 participants shows a 13.5% reduction in manual annotation actions, a 46.67% event match rate, and zero confirmed-field violations under the studied protocol. The code will be made publicly available at https://github.com/541741106/IMPACT_HOI.