AnkleType: A Hands- and Eyes-free Foot-based Text Entry Technique in Virtual Reality

Paper Title

AnkleType: A Hands- and Eyes-free Foot-based Text Entry Technique in Virtual Reality

Publication Info

• Topic area: Foot-based text entry techniques for Virtual Reality (VR)
• Keywords: Virtual Reality, text entry, hands-free, eye-free, foot-based interaction, ankle gestures, user study, keyboard layout, longitudinal evaluation, ergonomic design

Background and Problem

• Problem / challenge: Conventional VR text entry methods (e.g., physical keyboards, handheld controllers, gaze-based input) are limited when users' hands or visual attention are occupied. Existing foot-based methods require visual attention and wide foot movements, and primarily focus on sitting postures.
• Significance: Efficient and comfortable text entry in VR is critical for immersive applications like gaming, creativity, and collaboration. A hands- and eyes-free solution could enhance usability in scenarios where hands and eyes are engaged.
• Motivation and related work: Prior research explored hand-held controllers, gaze, and foot-based input for VR text entry. However, these approaches face challenges such as motion sickness, eye fatigue, and limited support for standing postures. This paper builds on foot-based input concepts and aims to address these gaps by enabling eye-free typing in both sitting and standing postures.

Solution

• Proposed approach: AnkleType, a foot-based text entry technique that uses ankle gestures for hands- and eyes-free typing in VR, supporting both standing and sitting postures.
• Novelty:
  1. Introduces a novel ankle-based typing system for hands- and eyes-free VR text entry.
  2. Conducts elicitation studies to identify user-preferred ankle gestures for text entry tasks.
  3. Optimizes keyboard layouts for eye-free input using user studies and a language model.
  4. Demonstrates learnability and effectiveness through a 7-day longitudinal study.
• Procedure and key techniques:
  1. Preliminary studies to measure ankle rotation ranges and elicit user-preferred gestures.
  2. Design of separate keyboard layouts for standing and sitting postures based on spatial and language modeling.
  3. Evaluation of unipedal and bipedal input strategies to identify optimal configurations for each posture.
  4. Longitudinal study to assess learning effects and eye-free typing performance.

Results

• Concrete findings:
  • Initial typing speeds: 8.99 WPM (BPSit) and 9.13 WPM (UPStand) for first-time users.
  • After 7 days of training:
    • Visual condition: 15.05 WPM (UPStand), 16.70 WPM (BPSit).
    • Eye-free condition: 11.15 WPM (UPStand), 12.87 WPM (BPSit).
    • Error rates decreased to 9.91% (UPStand) and 8.80% (BPSit) in the eye-free condition.
• Advantage over baselines:
  • Competitive performance compared to state-of-the-art hand-free and eye-free VR text entry methods.
  • Supports both standing and sitting postures, unlike prior works focused on a single posture.
• Experiments / evaluation:
  • Preliminary studies: 22 participants for ankle rotation range, 18 participants for gesture elicitation.
  • Keyboard layout optimization: 12 participants, spatial and language modeling.
  • Input strategy comparison: 16 participants, unipedal vs. bipedal strategies.
  • Longitudinal study: 12 participants over 7 days, visual and eye-free conditions.
• Limitations and future work:
  • Did not account for left-footed users; future work could explore adaptive designs.
  • Limited exploration of split-keyboard designs for bipedal input.
  • No evaluation in realistic dual-task or MR scenarios; future studies could address these contexts.

Summary

AnkleType introduces a novel hands- and eyes-free foot-based text entry system for VR, leveraging ankle gestures to support both standing and sitting postures. Through extensive user studies, it identifies optimal input strategies (UPStand for standing, BPSit for sitting) and achieves competitive performance with state-of-the-art methods. A 7-day longitudinal study demonstrates its learnability and effectiveness, with significant improvements in typing speed and accuracy under both visual and eye-free conditions. Future work could address left-footed users, split-keyboard designs, and evaluation in realistic scenarios. AnkleType offers a promising alternative for text entry in immersive environments.