BezelGlide: Interacting with Graphs on Smartwatches with Minimal Screen Occlusion

Title of the Paper

BezelGlide: Interacting with Graphs on Smartwatches with Minimal Screen Occlusion

Paper Information

• Research Area: Human-Computer Interaction, Data Visualization and Interaction Techniques for Smartwatches
• Keywords: Smartwatch, Data Graphs, Visualization, Interaction, Input

Research Background and Problem Statement

• Identified Problems or Challenges:

  1. The small screen size of smartwatches makes interaction with data visualizations challenging.
  2. Common touch interaction methods suffer from the "fat finger" problem, leading to poor touchpoint accuracy.
  3. Screen occlusion often disrupts users' ability to view and interact with data.

• Significance: Smartwatches are increasingly popular due to their convenience and wearability, especially for collecting health data. Designing effective data exploration and interaction techniques is crucial to unlocking the value of hidden data.

• Research Motivation and Related Work:

  • Current data visualization techniques on small-screen devices are limited; for example, traditional touch techniques are hindered by screen occlusion during interactions.
  • While some interaction techniques (e.g., Shift) address these issues to some extent, there is a lack of solutions optimized for data visualization, particularly in mobile and dynamic usage scenarios.

Proposed Solution

• Proposed Solution: The authors propose a novel interaction technique suite for smartwatches, called BezelGlide, specifically designed to optimize interactions with data graphs. It includes two variants:

  1. Full BezelGlide (FBG): Utilizes the entire bezel of the smartwatch for interaction.
  2. Partial BezelGlide (PBG): Uses only the bezel areas with minimal screen occlusion.

• Innovations:

  1. Leverages bezel-based interaction to reduce screen occlusion issues.
  2. Employs precise algorithmic design of interaction areas to accommodate the unique structure of circular smartwatch screens.
  3. Enables precise selection and viewing of data points without directly touching the screen.

• Implementation Steps and Key Techniques:

  1. Screen Occlusion Analysis (Study 1): Divided the smartwatch bezel into 24 equal segments and used a head-mounted camera to capture screen occlusion during user interactions, quantifying the degree of occlusion.
  2. Interaction Technique Design: Designed FBG and PBG techniques based on the occlusion study results. FBG uses the entire bezel, while PBG focuses on areas with minimal occlusion.
  3. Technique Evaluation (Study 2): Compared the techniques with the common Shift method under static and walking conditions, assessing performance metrics such as response time, interaction distance, and user preference.

Research Findings

• Specific Findings:

  1. FBG and PBG significantly alleviated the "fat finger" problem and effectively reduced screen occlusion.
  2. In data value detection tasks, PBG outperformed both FBG and the Shift technique.
  3. User studies confirmed that PBG was the most preferred method, primarily due to its minimal screen occlusion and more efficient interaction.

• Advantages over Existing Solutions: PBG reduced screen occlusion by approximately 34% compared to the Shift technique, while also requiring shorter finger movement distances, thereby improving task efficiency.

• Experimental or Evaluation Results:

  1. Average Screen Visibility: PBG achieved up to 90% visibility in optimal regions.
  2. Response Time: PBG had the lowest response time (approximately 1920ms), significantly outperforming FBG and Shift.
  3. Reduced Interaction Distance: PBG significantly shortened touchpoint movement paths, making it suitable for dynamic conditions.

• Limitations and Future Directions:

  • Limitations:

    1. Screen occlusion experiments used a head-mounted camera to capture the viewpoint, which may slightly differ from the user's actual perspective.
    2. The technique was designed for smartwatches worn on the left wrist, without considering the needs of left-handed users.
    3. Experiments were conducted in controlled laboratory conditions (e.g., walking on a treadmill), lacking validation in real-world dynamic scenarios.

  • Future Directions:

    1. Extend research to other data exploration tasks (e.g., data comparison, trend detection).
    2. Test the technique's performance in real-world scenarios (e.g., outdoor running).
    3. Adapt the technique for different smartwatch wearing styles (left wrist/right wrist).
    4. Explore integration with other key elements of smartwatch user interfaces (e.g., text content, icons).

Conclusion

The BezelGlide technique provides an innovative solution for interacting with data graphs on smartwatches, significantly mitigating screen occlusion issues through bezel-based interaction. Particularly, the PBG technique demonstrated superior task efficiency and user preference in both static and dynamic scenarios, showcasing strong design potential. However, further research is needed to adapt the technique to other scenarios and user conditions.