Andrea Zimmerer, M.Sc.

Abstract: Virtual reality (VR) exercise applications are promising tools, e.g., for at-home training and rehabilitation. However, existing applications vary significantly in key design choices such as environments, embodiment, and virtual coaching, making it difficult to derive clear design guidelines. A prominent design choice is the use of embodied virtual coaches, which guide user interaction and provide feedback. In a user study with 76 participants, we investigated how different levels of performance specificity in feedback from an embodied virtual coach affect intermediate factors, such as VR experience, motivation, and coach perception. Participants performed lower-body movement exercises, i.e., Leg Raises and Knee Extensions, commonly used in knee rehabilitation. We found that highly performance-specific feedback led to higher scores compared to medium specificity for perceived realism, as well as the anthropomorphism and sympathy of the virtual coach, but did not affect motivation. Based on our findings, we propose the design suggestion to include precise, performance-specific details when creating feedback for a virtual coach. We observed a descriptive pattern of higher scores in the low specificity condition compared to the medium condition on most measures, which raises the possibility that less specific feedback may, in some cases, be perceived more positively than moderately specific feedback. These findings provide valuable insights into how design choices impact relevant intermediate factors that are crucial for maximizing both workout effectiveness and the quality of the virtual coaching experience.

Abstract: Virtual reality applications employing avatar embodiment typically use virtual mirrors to allow users to perceive their digital selves not only from a first-person perspective but also from a holistic third-person view. However, due to distance-related biases such as the distance compression effect or a reduced relative rendering resolution, the self-observation distance (SOD) between the user and the virtual mirror might influence how users perceive their embodied avatar. Our article systematically investigates the effects of a short (1 meter), middle (2.5 meter), and far (4 meter) SOD between user and mirror on the perception of personalized and self-embodied avatars. The avatars were photorealistic reconstructed using state-of-the-art photogrammetric methods. Thirty participants were repeatedly exposed to their real-time animated self-embodied avatars in each of the three SOD conditions. In each condition, the personalized avatars were repeatedly altered in their body weight, and participants were asked to judge the (1) sense of embodiment, (2) body weight perception, and (3) affective appraisal towards their avatar. We found that the different SODs are unlikely to influence any of our measures except for the perceived body weight estimation difficulty. Here, the participants judged the difficulty significantly higher for the farthest SOD. We further found that the participants' self-esteem significantly impacted their ability to modify their avatar's body weight to their current body weight and that it positively correlated with the perceived attractiveness of the avatar. Additionally, the participants' concerns about their body shape affected how eerie they perceived their avatars. Both measures influenced the perceived body weight estimation difficulty. For practical application, we conclude that the virtual mirror in embodiment scenarios can be freely placed and varied at a distance of one to four meters from the user without expecting major effects on the perception of the avatar.

Abstract: ABSTRACT The development of embodied Virtual Reality (VR) systems involves multiple central design choices. These design choices affect the user perception and therefore require thorough consideration. This article reports on two user studies investigating the influence of common design choices on relevant intermediate factors (sense of embodiment, presence, motivation, activation, and task load) in a VR application for physical exercises. The first study manipulated the avatar fidelity (abstract, partial body vs. anthropomorphic, full-body) and the environment (with vs. without mirror). The second study manipulated the avatar type (healthy vs. injured) and the environment type (beach vs. hospital) and, hence, the avatar-environment congruence. The full-body avatar significantly increased the sense of embodiment and decreased mental demand. Interestingly, the mirror did not influence the dependent variables. The injured avatar significantly increased the temporal demand. The beach environment significantly reduced the tense activation. On the beach, participants felt more present in the incongruent condition embodying the injured avatar.

Abstract: Realistic and lifelike 3D-reconstruction of virtual humans has various exciting and important use cases. Our and others' appearances have notable effects on ourselves and our interaction partners in virtual environments, e.g., on acceptance, preference, trust, believability, behavior (the Proteus effect), and more. Today, multiple approaches for the 3D-reconstruction of virtual humans exist. They significantly vary in terms of the degree of achievable realism, the technical complexities, and finally, the overall reconstruction costs involved. This article compares two 3D-reconstruction approaches with very different hardware requirements. The high-cost solution uses a typical complex and elaborated camera rig consisting of 94 digital single-lens reflex (DSLR) cameras. The recently developed low-cost solution uses a smartphone camera to create videos that capture multiple views of a person. Both methods use photogrammetric reconstruction and template fitting with the same template model and differ in their adaptation to the method-specific input material. Each method generates high-quality virtual humans ready to be processed, animated, and rendered by standard XR simulation and game engines such as Unreal or Unity. We compare the results of the two 3D-reconstruction methods in an immersive virtual environment against each other in a user study. Our results indicate that the virtual humans from the low-cost approach are perceived similarly to those from the high-cost approach regarding the perceived similarity to the original, human-likeness, beauty, and uncanniness, despite significant differences in the objectively measured quality. The perceived feeling of change of the own body was higher for the low-cost virtual humans. Quality differences were perceived more strongly for one's own body than for other virtual humans.

Abstract: The illusion of virtual body ownership (VBO) plays a critical role in virtual reality (VR). VR applications provide a broad design space which includes contextual aspects of the virtual surroundings as well as user-driven deliberate choices of their appearance in VR potentially influencing VBO and other well-known effects of VR. We propose a protocol for an experiment to investigate the influence of deliberateness and context-match on VBO and presence. In a first study, we found significant interactions with the environment. Based on our results we derive recommendations for future experiments.

Abstract: This paper introduces an automated 3D-reconstruction method for generating high-quality virtual humans from monocular smartphone cameras. The input of our approach are two video clips, one capturing the whole body and the other providing detailed close-ups of head and face. Optical flow analysis and sharpness estimation select individual frames, from which two dense point clouds for the body and head are computed using multi-view reconstruction. Automatically detected landmarks guide the fitting of a virtual human body template to these point clouds, thereby reconstructing the geometry. A graph-cut stitching approach reconstructs a detailed texture. Our results are compared to existing low-cost monocular approaches as well as to expensive multi-camera scan rigs. We achieve visually convincing reconstructions that are almost on par with complex camera rigs while surpassing similar low-cost approaches. The generated high-quality avatars are ready to be processed, animated, and rendered by standard XR simulation and game engines such as Unreal or Unity

Abstract: This article investigates performance and user experience in Social Virtual Reality (SVR) targeting distributed, embodied, and immersive, face-to-face encounters. We demonstrate the close relationship between scalability, reproduction accuracy, and the resulting performance characteristics, as well as the impact of these characteristics on users co-located with larger groups of embodied virtual others. System scalability provides a variable number of co-located avatars and AI-controlled agents with a variety of different appearances, including realistic-looking virtual humans generated from photogrammetry scans. The article reports on how to meet the requirements of embodied SVR with today\u0027s technical off-the-shelf solutions and what to expect regarding features, performance, and potential limitations. Special care has been taken to achieve low latencies and sufficient frame rates necessary for reliable communication of embodied social signals. We propose a hybrid evaluation approach which coherently relates results from technical benchmarks to subjective ratings and which confirms required performance characteristics for the target scenario of larger distributed groups. A user-study reveals positive effects of an increasing number of co-located social companions on the quality of experience of virtual worlds, i.e., on presence, possibility of interaction, and co-presence. It also shows that variety in avatar/agent appearance might increase eeriness but might also stimulate an increased interest of participants about the environment.

Abstract: Culture, amongst other individual and social factors, plays a crucial role in human-human interactions. If robots should become a part of our society, they should be able to act in culture-specific manners as well. In this paper, we showcase the implementation of a cultural dichotomy, namely individualism vs. collectivism, in a social robots\u0027 conversation. Presenting these conversations to human observers from Germany and Japan, we investigate whether the implemented differences are recognized as such, and whether stereotypical culture-specific behaviours that correspond to the observers\u0027 cultural background is preferred. Results suggest that the manipulations in behaviour had the intended effect, but are not reflected in personal preferences.

Abstract: The demographic change causes an imbalance between the number of elderly in need of support and the number of caring staff. Therefore, it is important to help older adults keep their independence. Forgetting is a common obstacle people have to face when they become older which can be moderated by social robots by reminding on tasks. Since most elderly people are not used to robots a challenge in HRI is to identify aspects of a robot’s design to promote its acceptance. We present two different personas (companion vs. assistant) for a robotic platform by manipulating verbal and nonverbal behavior. A study was conducted in assisted living accommodations with the robot reminding on appointments to review if the persona influences the robot’s acceptance. Results indicate that the companion version of the robot was better accepted and perceived more likeable and intelligent compared to the assistant version.