IMPACT-Scribe: Interactive Temporal Action Segmentation with Boundary Scribbles and Query Planning
This work addresses the labor-intensive problem of dense temporal annotation for procedural activity videos by making the annotation process more efficient and interactive, though it is an incremental improvement over existing tools.
IMPACT-Scribe introduces a correction-driven framework for dense temporal action segmentation that reuses annotator corrections to improve future human-machine collaboration, achieving better labeling quality per effort and boundary accuracy in experiments and a human study.
Dense temporal annotation of procedural activity videos is vital for action understanding and embodied intelligence but remains labor-intensive due to reactive tools. Each correction is treated as an isolated edit, limiting reuse of information on annotator uncertainty and model reliability. We introduce IMPACT-Scribe, a correction-driven framework for dense labeling that uses each correction to improve future human-machine collaboration. IMPACT-Scribe combines uncertainty-aware boundary scribble supervision, local proposal modeling, cost-aware query planning, structured propagation, and correction-driven adaptation. Experiments and a human study show that this closed-loop design improves labeling quality per effort, enhances boundary accuracy, and fosters better human-machine interaction over time. The code will be made publicly available at https://github.com/BanzQians/IMPACT_AS.