Author:

Tim Brüggemann
Supervisor:Prof. Gudrun Klinker, Ph.D.
Advisor:Linda Rudolph, M.Sc.
Submission Date:15.01.2024

Abstract

The goal of this thesis is to examine and compare two data visualization methods for diagrams, namely a traditional 2D representation of a diagram on the screen of a mobile device on the one hand and on the other hand the placement of an image plane containing the diagram as an AR-element in a 3D immersive scene. Therefore, a prototype was developed to examine the feasibility of the two methods, both in regards to technical realization and user acceptance.

As a result of the subsequent user study, a linked 2D and 3D approach is recommended. The AR visualization provides great overall orientation and navigation, while the 2D visualization offers more precise manipulation and spatial flexibility. The main hurdle in this comparison is the fact that the 2D visualization is already well-established and ever-present in our daily lives, which influenced the opinions of many participants. With the shift in demographics and children coming in contact with AR at much earlier ages, this remains an interesting topic in need of further investigation in the future

Conclusion

The results of this user study are not clear enough to draw meaningful conclusions about which visualization method is better. The study acts more as a starting point to the problem statement, discovering which aspects work and which do not through several early pilot studies and a formative evaluation, disclosing the opinions of a broad demographic group. Moreover, while the results of the UEQ mostly allow no precise comparisons - partially due to the discussed limitations, partially due to the early stage in production - both visualization methods have received overall high scores of acceptance. The question of which method is better is a complex topic with many facets. Currently, both methods prevail in certain areas, and the usefulness of each method is highly dependent on the given use case and environment. Comparing more specific aspects of the two methods requires an improvement of the World Space method according to the feedback from this study and further focused user studies with an appropriate range and selection of demographic groups.

With immersive display technologies constantly improving, some identified problems of the World Space method, such as unsteady tracking, might be solved automatically in a few years. According to Mark Weiser's vision of ubiquitous computing, combined with the possibilities of Situated Analytics and the rise of mixed reality technologies, it is highly likely that AR will be used for the presentation of data, even if it should be revealed that this is not the most suitable medium for this purpose. Therefore, continued research on these kinds of visualization techniques is essential to develop evidence-based guidelines on how to best integrate these technologies into the visualization process.

As a result of this user study, a linked 2D and 3D approach is recommended. World Space provides great overall orientation and navigation, while Screen Space offers more precise manipulation and spatial flexibility. A possible way to realize this combination could be to offer a way (e.g., a button) to switch dynamically between the two visualization methods. Switching from World Space to Screen Space projects the image plane from world coordinate space onto screen coordinate space and smoothly resets the rotation. The user can use all interaction methods of Screen Space until switching back to World Space, which seamlessly projects the image into the world scene. An alternative to a SwitchMethod button could be to have World Space be the default method and only temporarily change to Screen Space while the user performs the swiping or pinching gestures.

PDF Thesis

GitLab Repository

https://gitlab.lrz.de/IN-FAR/Thesis-Projects/ba-ws2324-tim-brueggemann-comparing-diagrams