UbiTouch: Towards a Universal Touch Interface
Dev Shah
TL;DR
This work targets ubiquitous computing by enabling touch interactions on arbitrary surfaces using a single thermal camera. The core approach, UbiTouch, combines a two-pass ROI-driven touch-detection pipeline with lightweight, vision-based processing to differentiate hover from touch, support multi-finger inputs, and reconstruct trajectories, while also addressing jitter through sensor fusion. Key contributions include a practical thermal-imaging–based framework that minimizes computational load and hardware requirements, enabling scalable interaction in diverse environments, with promising preliminary accuracy. The work lays groundwork for on-device, surface-agnostic touch sensing and points to future enhancements via RGB cues, improved hand detection, and large-scale testing to bolster robustness and applicability in real-world settings.
Abstract
Touch is one of the most intuitive ways for humans to interact with the world, and as we advance toward a ubiquitous computing environment where technology seamlessly integrates into daily life, natural interaction methods are essential. This paper introduces UbiTouch, a system leveraging thermal imaging to detect touch interactions on arbitrary surfaces. By employing a single thermal camera, UbiTouch differentiates between hovering and touch, detects multi-finger input, and completes trajectory tracking. Our approach emphasizes the use of lightweight, low-computation algorithms that maintain robust detection accuracy through innovative vision-based processing. UbiTouch aims to enable scalable, sustainable, and adaptable interaction systems for diverse applications, particularly with regards to on-human sensing.