Smartwatches require efficient text entry methods, particularly for round face designs with limited screen space. This paper presents D-shaped EdgeWrite, a space-saving, stabilized text entry system for round face smartwatches. Inspired by the EdgeWrite technique, it employs a D-shaped input area in the lower-right screen region, facilitating edge-based unistroke gestures for enhanced accuracy. A thumb-supported posture further stabilizes input, improving usability while walking or multitasking. The system integrates template matching and a language model, resolving gesture ambiguities via character predictions. Comparative evaluations against ZoomBoard and round face EdgeWrite showed competitive text input speeds (5.76 WPM sitting, 5.61 WPM walking) and the lowest corrected error rates (3.81% sitting, 5.75% walking). User feedback highlights ease of use and customizable gestures as key advantages. While promising, the system is optimized for right-handed users and requires visual attention for candidate selection. Future work will explore left-handed adaptation and eyes-free input solutions. The D-shaped EdgeWrite system offers a practical alternative for smartwatch text entry, improving speed and accuracy in real-world conditions.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

D-Shaped EdgeWrite: A Space-Saving Stabilized Text Entry Method for Round Face Smartwatches

  • Kentaro Go,
  • Riho Hamanaka,
  • Keiichi Ueno,
  • Yuichiro Kinoshita

摘要

Smartwatches require efficient text entry methods, particularly for round face designs with limited screen space. This paper presents D-shaped EdgeWrite, a space-saving, stabilized text entry system for round face smartwatches. Inspired by the EdgeWrite technique, it employs a D-shaped input area in the lower-right screen region, facilitating edge-based unistroke gestures for enhanced accuracy. A thumb-supported posture further stabilizes input, improving usability while walking or multitasking. The system integrates template matching and a language model, resolving gesture ambiguities via character predictions. Comparative evaluations against ZoomBoard and round face EdgeWrite showed competitive text input speeds (5.76 WPM sitting, 5.61 WPM walking) and the lowest corrected error rates (3.81% sitting, 5.75% walking). User feedback highlights ease of use and customizable gestures as key advantages. While promising, the system is optimized for right-handed users and requires visual attention for candidate selection. Future work will explore left-handed adaptation and eyes-free input solutions. The D-shaped EdgeWrite system offers a practical alternative for smartwatch text entry, improving speed and accuracy in real-world conditions.