HeadZoom: Hands-Free Zooming and Panning for 2D Image Navigation Using Head Motion

2025-08-03

AI Summary

To address touchless 2D image navigation—such as map browsing and medical image review—this paper proposes a VR-based head-controlled interaction method leveraging real-time head pose estimation. Our core contribution is Parallel Zoom, a novel navigation mode that decouples translational and zooming head motions along orthogonal axes, extending conventional Static (trigger-based) and Tilt Zoom (tilt-driven) paradigms. This decoupling significantly reduces required head movement amplitude and associated user fatigue. A user study demonstrates that Parallel Zoom outperforms baseline methods across key metrics: task completion time improves by 23.6%, error rate decreases by 31.4%, and subjective comfort scores increase significantly (p < 0.01). The gains are especially pronounced for prolonged, high-precision spatial tasks. Overall, this work establishes an efficient, low-cognitive-load, and intuitively natural interaction paradigm for contactless image navigation.

Technology Category

• Computer Vision: Multi-modal Vision
• Intelligent Robots: Localization, Mapping, and Navigation
• Humans and AI: Interaction Techniques and Devices

Application Category

• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Security and Privacy: Large-scale security measurements

Abstract

We introduce extit{HeadZoom}, a hands-free interaction technique for navigating two-dimensional visual content using head movements. The system enables fluid zooming and panning by only using real-time head tracking. It supports natural control in applications such as map exploration, radiograph inspection, and image browsing, particularly where physical interaction is limited. We evaluated HeadZoom in a within-subjects user study comparing three interaction techniques-Static, Tilt Zoom, and Parallel Zoom-across spatial, error, and subjective metrics. Results show that Parallel Zoom significantly reduced total head movement compared to Static and Tilt modes. Users reported significantly lower perceived exertion for Parallel Zoom, confirming its suitability for prolonged or precision-based tasks. By minimising movement demands while maintaining task effectiveness, HeadZoom advances the design of head-based 2D interaction in VR, creating new opportunities for immersive, accessible, and hands-free systems for image exploration.

Problem

• Hands-free 2D image navigation using head motion
• Reducing head movement for prolonged precision tasks
• Enhancing VR interaction for accessible image exploration

Innovation

• Hands-free zooming and panning via head motion
• Real-time head tracking for fluid navigation
• Parallel Zoom reduces movement and exertion