Project overview
Our shared-attention work runs in a Unity environment on head-mounted displays, where each participant sees the other as an avatar. The CAVE and the PowerWall at the LRZ put the same virtual scene in a room where people keep their own bodies and can see each other directly. This project rebuilds the environment in Unreal so that it runs on those installations, and then uses it for a study of your own design.
Project motivation
Everything the lab has measured about how one person reads another’s attention has been measured through an avatar, because a headset leaves no other option. A projection-based room removes that constraint: the scene is still virtual, but the bodies in it are real. Rooms of this kind have been built for research since the original CAVE [1], and how much immersion a system provides is known to change how people perform and behave [2], so the display deserves to be treated as a variable rather than a fixed cost.
The port is not incidental work. These rooms are driven by synchronised render clusters rather than by a single headset, and Unreal’s nDisplay framework targets that class of hardware directly, which is why rebuilding is the honest route rather than stretching the existing Unity project to fit. The move is also a measurement in itself: what survives it and what has to be built again says a great deal about which parts of an experiment were load-bearing and which were only ever artefacts of the engine.
Project goals
- The port. Rebuild the environment in Unreal and get it running on the CAVE and the PowerWall at the LRZ: scene, interaction, tracking integration and data logging, with stereo and rendering verified across the display cluster, and with the logged output in the format the existing analysis tooling already reads.
- The study. Design and run a study in the ported environment. The question is yours to choose, within the lab’s line of work on shared attention and within what the rooms can support. Comparing the same task in a headset against the CAVE, using the real bodies a CAVE affords to test cues an avatar cannot carry, or using the PowerWall for configurations with an audience rather than a partner are all reasonable starting points. Agree the question with your supervisors before building conditions around it.
You will
- Perform a literature review on projection-based virtual reality and display fidelity
- Rebuild the existing Unity environment in Unreal for the CAVE and the PowerWall
- Integrate tracking and logging so the recorded data matches the existing pipeline
- Propose a study question, settle it with your supervisors, and run it
- Analyze the data with the existing Python/R mixed-model tooling
- Summarize your findings in a thesis and present them to an audience
- (Optional) co-write a research paper
You need
- Strong communication skills in English
- Good knowledge of Unreal, or solid Unity experience and the appetite to change engines
- Comfort with C++ or Blueprints
- Interest in experimental design, with basic statistics a plus
- Willingness to work on site at the LRZ in Garching, where the CAVE and the PowerWall are
Co-supervised with Elisabeth Mayer, Leibniz Supercomputing Centre (LRZ).
References
- Cruz-Neira, C., Sandin, D. J., & DeFanti, T. A. (1993). Surround-screen projection-based virtual reality: the design and implementation of the CAVE. SIGGRAPH ’93, 135–142. https://doi.org/10.1145/166117.166134
- Bowman, D. A., & McMahan, R. P. (2007). Virtual Reality: How Much Immersion Is Enough? Computer, 40(7), 36–43. https://doi.org/10.1109/MC.2007.257
- The parent study on avatar-mediated shared attention in virtual reality; details are available from the advisor on request.
Keywords: Virtual Reality, CAVE, PowerWall, Unreal Engine, Shared Attention, Display Fidelity