TangibleNet: Synchronous Network Data Storytelling through Tangible Interactions in Augmented Reality
Authors
Research Background and Problem
-
Identified Problems or Challenges
The authors identified that real-time dynamic network data storytelling (synchronous data storytelling) still faces significant challenges, particularly in performing complex operations on network components during live presentations. Existing research primarily focuses on asynchronous scenarios, while tools for real-time data presentation remain scarce. In synchronous data storytelling, presenters need to dynamically manipulate various elements in node-link diagrams (e.g., nodes, links, and annotations). Relying solely on gestures leads to high learning costs, increased error rates, and difficulty in memorization. Therefore, there is a need to explore more natural and intuitive interaction methods. -
Significance
Real-time data storytelling is widely used in contexts such as organizational decision-making meetings and news reporting. Successful synchronous storytelling requires combining narrative content with visual aids and dynamically engaging the audience's attention. This capability not only supports real-time decision-making but also enhances data communication to improve learning and cognition. -
Research Motivation and Related Work
Starting from observations of news anchors and feedback from HCI/VIS researchers, the authors aim to leverage the advantages of physical object interaction to simplify the manipulation of graphical elements during live presentations. Previous studies have explored gestures and augmented reality systems, but these studies provide insufficient support for network data storytelling. Moreover, physical interaction is considered to enhance comprehension and interaction satisfaction, highlighting the need for an interaction design framework tailored to dynamic network diagrams.
Solution
-
Proposed Method or Solution
The authors propose an augmented reality (AR)-based system called TangibleNet, which allows presenters to interact with node-link diagrams using double-sided magnets. This hybrid approach, combining physical interaction with digital augmented reality, offers a novel method for synchronous network data storytelling. -
Innovations
- Introduction of Physical Object Interaction: Compared to traditional gesture-driven or multi-touch display devices, the use of magnets enables more intuitive operations, granting users greater flexibility and agency.
- Interaction Design Space: A three-dimensional mapping between user actions and network operations was defined through workshops, encompassing interaction commands, primary interaction modes (e.g., sliding, overlaying), and multi-object operability (single-object or multi-object collaborative operations).
- Dynamic Mapping and Registration Mechanism: The system supports real-time registration of physical objects with nodes, allowing presenters to adjust content on the fly without complex pre-configuration.
-
Implementation Steps and Key Technologies
- Conducted interviews with news anchors and a workshop with 14 participants to identify system requirements and potential physical interactions. Collected user feedback on the design of command mappings and summarized the design space using classification dimensions.
- Developed a TangibleNet prototype system, integrating computer vision technologies (OpenCV and MediaPipe) to track magnets and gestures, and built an interactive visualization environment using projection and a whiteboard.
- Consolidated and refined the integration of user-performed physical actions and system responses during experiments, reducing interaction errors and improving the fluidity of the experience.
Research Outcomes
-
Specific Outcomes
- System Prototype Development: TangibleNet was implemented, enabling real-time data storytelling through the combination of double-sided magnets and gestures in augmented reality.
- Definition of Design Space: The mapping of visual commands to physical actions was incorporated into a three-dimensional framework (interaction commands, primary interaction modes, multi-object operability), providing theoretical support for future system designs.
- User Research and Experimental Evaluation: Experiments with 12 participants demonstrated that the interaction method was easy to learn, provided presenters with greater autonomy, and showed potential for application in real-time network visualization storytelling.
-
Advantages Compared to Existing Methods
- TangibleNet is the first to empower synchronous data storytelling through physical interaction, significantly simplifying the operation of complex visual commands.
- High autonomy: Presenters can dynamically adjust, store, display, and hide nodes or links at any time, rather than relying on pre-configured slides.
- Multi-modal interaction enhances flexibility, combining gestures with physical objects.
-
Experimental Results
User feedback indicated that the system has an intuitive learning curve, with most participants becoming familiar with the interaction method in a short time. Additionally, users found the system's operations to be highly coherent with its visual effects, facilitating real-time adjustments to better respond to audience questions. -
Limitations and Future Directions
- Limited Network Commands: The current prototype does not fully support all possible operations in dynamic network scenarios (e.g., filtering nodes).
- Technical Constraints: Vision-based recognition using computer vision is sensitive to lighting conditions and occlusion, which may affect accuracy in certain environments.
- Non-Remote Application: The system is currently designed for face-to-face storytelling scenarios, and future work should explore interactive features for remote audiences.
- Expansion of Physical Objects: While double-sided magnets provide an intuitive design, other physical interaction objects (e.g., shape-changing props) could express richer semantics in specific contexts and warrant further exploration.
Through TangibleNet, this paper addresses the technological gap in synchronous network data storytelling and provides a potential reference for future applications of augmented reality technology to enhance interaction experiences in education, television broadcasting, and commercial environments.
Research Questions / Practical Problems
Question signals indexed for this paper.
Research Questions
3- How can physical object interaction enhance naturalness and intuitiveness of real-time dynamic network data storytelling?Category: XR Input, Tracking, and Spatial InteractionSimilar questionsarrow_forward
- Can dual-sided magnets combined with AR simplify real-time manipulation of complex network graphs?Category: XR Input, Tracking, and Spatial InteractionSimilar questionsarrow_forward
- What interaction design framework suits real-time presentation needs of dynamic network graphs?Category: XR Input, Tracking, and Spatial InteractionSimilar questionsarrow_forward
Practical Problems
1- Presenters easily make errors and struggle to manipulate dynamic network graphs intuitively in real-time data presentations.Category: XR Input, Tracking, and Spatial InteractionSimilar questionsarrow_forward
- 80%
Chameleon: Bringing Interactivity to Static Digital Documents
CHI '20· Interactive Data Visualization +1
- 67%
Answering Questions about Charts and Generating Visual Explanations
CHI '20· Interactive Data Visualization +1
- 67%
IGScript: An Interaction Grammar for Scientific Data Presentation
CHI '21· Interactive Data Visualization +1
Based on Jaccard similarity of research subtopics & professions (≥60%)