Cross, Dwell, or Pinch: Designing and Evaluating Around-Device Selection Methods for Unmodified Smartwatches

TL;DR

Cross, Dwell, or Pinch investigates around-device input on unmodified smartwatches via a one-dimensional sonar-based finger-tracking system (SonarSelect). It compares three selection triggers across binary and multi-target tasks, finding Double-crossing fastest for binary selections and Dwelling most effective for consecutive multi-target tasks, with haptic feedback enhancing comfort but not performance. The work demonstrates the practical viability of sonar-based cursor control on commodity hardware and provides design guidance for around-device smartwatch interfaces, while releasing the system as open-source to spur further research. Overall, the study offers concrete performance benchmarks and actionable recommendations for implementing around-device selection on current smartwatch platforms.

Abstract

Smartwatches offer powerful features, but their small touchscreens limit the expressiveness of the input that can be achieved. To address this issue, we present, and open-source, the first sonar-based around-device input on an unmodified consumer smartwatch. We achieve this using a fine-grained, one-dimensional sonar-based finger-tracking system. In addition, we use this system to investigate the fundamental issue of how to trigger selections during around-device smartwatch input through two studies. The first examines the methods of double-crossing, dwell, and finger tap in a binary task, while the second considers a subset of these designs in a multi-target task and in the presence and absence of haptic feedback. Results showed double-crossing was optimal for binary tasks, while dwell excelled in multi-target scenarios, and haptic feedback enhanced comfort but not performance. These findings offer design insights for future around-device smartwatch interfaces that can be directly deployed on today's consumer hardware.

Figures (7)

• Figure 1: Prior work has proposed various smartwatch systems featuring edge-located uni-dimensional input to avoid occlusion issues during interaction. For example, minimizing the input space by using the edge of the screen EdgeInteraction18IJHCSAhn or the touching force applied to a corner 1DForceFitts23Ren, or using physical controls, such as a bezel bezel20Wongbezel22Rey, crown CrownBoard23Rakhmetulla, or case CaseTouch24MUM
• Figure 2: A) SonarSelect evaluation setup: a smartwatch worn on a prop forearm and a prop finger mounted on a linear stage that moves left and right over the back of the prop hand and wrist. B) Mean movement distance error at different movement ranges captured using this setup and compared to a baseline implementation (LLAP LLAP16Wang). This data shows that SonarSelect shows strong performance for movements within 10cm of the smartwatch.
• Figure 3: Study interface and interaction. Two targets and a cursor are displayed on the smartwatch A). A participant controls the cursor with around-device finger movements B). The different hand postures adopted by participants in Study 1 C): Most participants (16/18) opted for on-skin interaction, where they positioned one or two fingers on the back of the hand wearing the watch, while the remaining two participants chose to interact in the air. These choices were not instructed or constrained.
• Figure 4: Usability questionnaires. Mean perceived comfort A) and workload B) of the three selection triggers.
• Figure 5: Study interface A) and haptic feedback design for target activation and confirming selection B).