A Design Space For Data Visualisation Transformations Between 2D And 3D In Mixed-Reality Environments

Title of the Paper

A Design Space For Data Visualisation Transformations Between 2D And 3D In Mixed-Reality Environments

Paper Information

• Domain: Data visualization and mixed-reality interaction design
• Keywords: Visualization, Mixed reality, Immersive analytics, Animated transformations, Direct manipulation

Research Background and Problem

• Problem and Challenges: The transformation of data visualization between 2D displays and 3D spaces remains underexplored. While existing studies have touched upon 2D and 3D collaboration, there is still a lack of systematic frameworks for designing interactive transformations in mixed-reality environments.
• Significance: With the advancement of mixed-reality technologies, the ability to flexibly switch between 2D and 3D visual environments for data representation is crucial for various user tasks such as data exploration and decision support. This capability enables users to dynamically configure data views based on their needs.
• Research Motivation: The authors aim to address the current gap in research on the design space for 2D-to-3D data visualization transformations, providing guidance for future design and development.

Solution

• Approach: The authors propose a detailed design space to define the fundamental framework for transforming data visualization between 2D and 3D (and vice versa) in mixed-reality environments.
• Innovations:
  • Introduced a classification of visual transformations and mapped them to specific user tasks.
  • Designed a comprehensive framework covering the transformation process from initial visualization states to final states, including user interaction, visual changes, and transformation control.
  • Provided examples of practical techniques to illustrate the application of the design space through real-world cases.
• Implementation Steps and Techniques:
  1. Define Transformation Possibilities: Map data visualization to four states (2D single view, 3D single view, 2D multi-view, 3D multi-view) to clarify transformation goals.
  2. Design User Interaction Modes: Define methods for users to trigger and control transformations using multimodal interaction devices (e.g., gestures, touch, voice).
  3. Implement Transformation Control: Offer continuous or staged animation transformation control, considering factors such as duration, persistence, and task relevance.
  4. Create Example Techniques: Examples include "expanded scatterplots" and "shortest path visualization" as specific transformation techniques.

Research Outcomes

• Specific Results:

  • Proposed and validated a design space framework to support dynamic transformations of data visualization between 2D and 3D.
  • Developed a series of representative transformation techniques, demonstrating how interactive animations can transition data from simple layouts to complex 3D data views.
  • Suggested general principles for future design implementations, such as reducing reliance on 3D and optimizing the intuitiveness of user interactions.

• Advantages:

  • The proposed approach organizes scattered 2D-to-3D transformation techniques into a unified framework.
  • Offers a novel perspective on interactive design, enabling users to dynamically choose suitable 2D/3D visual states based on specific data tasks.
  • Enhances user control and flexibility in data exploration compared to traditional animated transformations.

• Experimental Evaluation Results:

  • Example techniques were validated through a mixed-reality prototype on Microsoft HoloLens 2, demonstrating the operability and generalizability of the design space framework.

• Limitations and Future Directions:

  • Limitations:
    • Lack of user study data to support the practical utility of the proposed framework.
    • Example techniques are primarily based on planar display environments and have not been extended to non-Euclidean surfaces or other specialized scenarios.
  • Future Directions:
    • Explore chained and branched transformation possibilities, i.e., how multiple transformations can be combined into a composite interaction behavior.
    • Investigate the impact of different input modalities within the design space on user experience and propose optimized interaction structures.
    • Integrate user configuration tools into the design space, allowing users to customize visualization transformations to suit their specific task requirements.

By structuring the information from this paper, it provides a clear understanding of its core content and contributions, serving as a reference point for further in-depth research.