Vera Marie Memmesheimer

• Extended Reality (XR) unlocks new opportunities for interacting with spatial content, supporting both individual users and diverse collaborative settings. Different degrees of virtuality like Augmented/Mixed Reality (AR/MR) and Virtual Reality (VR) as well as different devices like head-mounted displays (HMDs) and handheld displays (HHDs) offer distinct benefits for different use cases. Despite the immense interest XR has sparked in numerous domains, it is rarely used in practice where scalability limitations outweigh XR's intrinsic potential. Since users are typically involved in multiple use cases, leveraging technology-specific benefits requires switching between use cases and XR technologies. However, existing user interfaces (UIs) impede these transitions because they are tailored to specific use cases. Thus, this dissertation is concerned with the development of highly scalable UIs which facilitate switching between XR applications that differ in technology and number of users. First, Scalable Extended Reality (XRS) is introduced as a novel concept for XR spaces which scale across different devices, degrees of virtuality, and varying numbers of potentially distributed users. A research agenda addressing the barriers to the realization of XRS is established based on an extensive compilation of related research. As an initial step towards XRS and as the basis for this dissertation the XRS framework is developed. By sharing spatial content among collaborators, XR could overcome a key limitation of conventional videoconferencing tools. However, collaboration support features face scalability issues when different XR technologies are used in large groups. A particular challenge concerns the accurate representation of HHD users. In response, the dissertation presents new insights from a detailed study on how humans interact with HHDs across different display sizes, display orientations, and body poses. Extending these results, mechanisms for individually activating the visibility of awareness cues are designed to reduce visual overload in large groups. Next, interaction techniques scaling with devices and degrees of virtuality are presented. Starting with MR-HHDs, a unified paradigm for object translation and rotation is developed and evaluated. By combining tablet movement and peripheral touch input its minimalist design overcomes key issues of prior methods. Extending HMDs with a tablet controller using this paradigm yields consistent interaction with MR-HHDs, MR-HMDs, and VR-HMDs. In a comparative study, this solution outperformed state-of-the-art methods and revealed high scalability. Eventually, the novel UIs are implemented and evaluated for robot control and factory layout planning, showcasing their practical applicability.