Immersive Contextual Data Analytics

Project Description

Recent advances in mixed reality (MR) technology, which seamlessly blend the physical environment with computer-generated content around the user, have reduced the barriers to practically storing and presenting a richer array of contextual information. Thanks to substantial investments from “tech giants” like Microsoft, Google, Meta, and Apple, mixed reality displays have rapidly evolved from limited, heavy, and computer-tethered devices to standalone, lightweight alternatives resembling conventional screens. Furthermore, technology to digitally capture and depict physical environments has also rapidly improved. The recent introduction of 3D Gaussian splatting techniques[1] has considerably reduced the cost and time required to create high-fidelity representations of 3D scenes from standard camera imagery. Such spatial computing applications represent the most significant paradigm shift in human-computer interaction (HCI) since the introduction of graphical user interfaces, providing opportunities for creating rich contextual environments without having to conduct site visits.

This PhD project aims to leverage innovative spatial computing technologies and proposes Immersive Contextual Data Analytics (ICDA) as a method to address contextual analysis challenges by bringing rich contextual information to the analyst’s workspace. Despite the technological capability to support ICDA, there remains a lack of fundamental human-computer interaction research and usability design principles to realise practical and effective applications, particularly concerning how data visual analytics translates to this new method. This concept is aligned with a recently-developed terminology and conceptual model for blended real and virtual data representations, namely proxsituated (proxy of situated) visualisation and analytics[2]. The concept of a proxy in this context refers to contextual information that serves as a representation or substitute for actual real-world data. This project will draw inspiration from research into how spatial awareness and context-specific memory function in mixed reality (MR) data analytics environments[3].

Expected outcomes

The anticipated results of this doctoral research include:

• a comprehensive framework for conceptualising techniques related to ICDA,
• empirically derived insights and knowledge about the application and effectiveness of ICDA methods.

References

[1] Chen, G., & Wang, W. (2024). A survey on 3d gaussian splatting. arXiv preprint arXiv:2401.03890.

[2] Satriadi, K. A., Cunningham, A., Smith, R. T., Dwyer, T., Drogemuller, A., & Thomas, B. H. (2023, April). Proxsituated visualization: An extended model of situated visualization using proxies for physical referents. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (pp. 1-20).

[3] Satriadi, K. A., Tag, B., & Dwyer, T. (2023). Context-Dependent Memory in Situated Visualization. arXiv preprint arXiv:2311.12288.

Scholarships

Scholarships may be provided for an excellent candidate.