Sticky-Glance: Robust Intent Recognition for Human Robot Collaboration via Single-Glance

2026-03-06

AI Summary

This work addresses the challenge of robust gaze-based intention recognition in dynamic multi-object environments, where performance is often degraded by noise, microsaccades, and viewpoint variations. To overcome these limitations, the authors propose an object-centric gaze anchoring framework featuring a novel sticky-glance algorithm that jointly models the geometric distance and directional trends between gaze points and candidate targets. This approach enables stable intention anchoring with as few as three gaze samples. Integrated with continuous shared control and a multimodal human-in-the-loop interaction mechanism, the system significantly outperforms baseline methods, achieving a dynamic target tracking rate of 0.94 and a static target selection accuracy of 0.98, while reducing task completion time by nearly 10%. These results demonstrate enhanced responsiveness and robustness in human-robot collaboration.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Computer Vision: Motion & Tracking

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Gaze is a valuable means of communication for impaired people with extremely limited motor capabilities. However, robust gaze-based intent recognition in multi-object environments is challenging due to gaze noise, micro-saccades, viewpoint changes, and dynamic objects. To address this, we propose an object-centric gaze grounding framework that stabilizes intent through a sticky-glance algorithm, jointly modeling geometric distance and direction trends. The inferred intent remains anchored to the object even under short glances with minimal 3 gaze samples, achieving a tracking rate of 0.94 for dynamic targets and selection accuracy of 0.98 for static targets. We further introduce a continuous shared control and multi-modal interaction paradigm, enabling high-readiness control and human-in-loop feedback, thereby reducing task duration for nearly 10 \%. Experiments across dynamic tracking, multi-perspective alignment, a baseline comparison, user studies, and ablation studies demonstrate improved robustness, efficiency, and reduced workload compared to representative baselines.

Problem

• gaze-based intent recognition
• multi-object environments
• gaze noise
• dynamic objects
• micro-saccades

Innovation

• sticky-glance
• gaze-based intent recognition
• object-centric grounding
• shared control
• multi-modal interaction