Can human observers predict which of two directions a car will take in an accident situation when watching videos before the event recorded from the driver's first-person perspective? And is it possible to use the...
详细信息
ISBN:
(纸本)9798331506926;9798331506919
Can human observers predict which of two directions a car will take in an accident situation when watching videos before the event recorded from the driver's first-person perspective? And is it possible to use the observers' eye-gaze data to predict their direction-choices? In our study with N = 30 participants, we first show that observers identify the correct direction already from 4 seconds(s) before the accident event with performance rising to 92% at 1s prior. Statistical analyses of the eye gaze data of the observers further identify patterns of gaze behaviors differentiating the observers' choices. We then use an explainability approach to show that graph networks pay attention to similar scene parts as humans. Our results showcase the remarkable ability of human action predictions and that these predictions during complex, dynamic viewing can be classified from gaze data alone.
When faced with learning new algorithms, students often struggle with recognizing the details of the algorithm or with understanding the meaning of its intermediate stages and interpreting the results. These difficult...
详细信息
ISBN:
(纸本)9798350374490;9798350374506
When faced with learning new algorithms, students often struggle with recognizing the details of the algorithm or with understanding the meaning of its intermediate stages and interpreting the results. These difficulties typically stem from the inherent inadequacies of using static representations to describe dynamic systems. We have developed the guided walk-through experience, Walk Me Through It, as a way for users to, not only visualize but interact, with every stage of common graph traversal algorithms, Breadth-First Search and Depth-First Search. Users must construct their own map of the room layout and connections between them by collecting room Tokens as they navigate the maze implemented with Flexible Spaces.
Adopting Artificial Intelligence (AI) in organizational settings creates a dichotomy between economic gains for stakeholders and job security concerns for employees, leading to documented avoidance behavior. This resi...
详细信息
ISBN:
(纸本)9798400704772
Adopting Artificial Intelligence (AI) in organizational settings creates a dichotomy between economic gains for stakeholders and job security concerns for employees, leading to documented avoidance behavior. This resistance can negatively impact innovation and market competitiveness. This paper examines perceived workload (PW), an area not yet fully explored, and it is a critical factor in employees' reluctance to engage with AI technologies. The research investigates how task delegation to AI influences PW and the subsequent avoidance behavior, considering the legal context that precludes AI accountability and necessitates human oversight. The study will incorporate responsibility denial (RD) as a cognitive mechanism within its theoretical framework, examining how and why RD among employees influences organizational outcomes and AI adoption. The findings will help identify how to effectively manage PW perceptions and could potentially reduce AI avoidance, facilitating better AI integration into workplaces and protecting corporate investments.
Gender identification of virtual reality (VR) users by machine learning tracking data could afford personalized experiences, including mitigation of expected human factors or psychology issues. While much research has...
详细信息
ISBN:
(纸本)9798350374490;9798350374506
Gender identification of virtual reality (VR) users by machine learning tracking data could afford personalized experiences, including mitigation of expected human factors or psychology issues. While much research has recently been conducted to identify individual users given their VR tracking data, little research has investigated gender identification. Furthermore, nearly all prior studies have only considered positions and rotations of all the devices. We present a systematic investigation of different combinations and spatial representations of VR tracked devices for predicting a user's gender. Our results indicate head rotations are integral to gender identification while head positions are surprisingly not as important.
In this paper, we suggest a foldable device that renders the shape and texture in Virtual Reality called Deep-Texture. We devised Deep-Texture to achieve effective haptic feedback with lightweight hardware by combinin...
详细信息
ISBN:
(纸本)9798350374490;9798350374506
In this paper, we suggest a foldable device that renders the shape and texture in Virtual Reality called Deep-Texture. We devised Deep-Texture to achieve effective haptic feedback with lightweight hardware by combining the basic sensations of shape and texture. By integrating the frequency change of linear resonant actuator and 1-bar mechanism, we propose a novel haptic interaction device for immersive VR experiences. During the pilot test, the results show that it enhances realism while keeping usability. By open-sourcing Deep-Texture, we aim to empower a wider range of users to engage with and benefit from haptic technology, ultimately lowering the hurdles.
Interactive agents in virtual reality (VR) are anticipated to make decisions and provide feedback based on the user's inputs. Despite recent advancements in large language models (LLMs), employing LLMs in real-tim...
详细信息
ISBN:
(纸本)9798350374490;9798350374506
Interactive agents in virtual reality (VR) are anticipated to make decisions and provide feedback based on the user's inputs. Despite recent advancements in large language models (LLMs), employing LLMs in real-time face-to-face interactions decision-making, and delivering personalized feedback in VR remains challenging. To address this, our proposed system involves generating and labeling symbolic data, pre-training a real-time network, collecting personalized data, and fine-tuning the network. Utilizing inputs such as interaction distances, head orientations, and hand poses, the agents can provide personalized feedback. User experiments show significant advantages in both pragmatic and hedonic aspects over LLM-based agents, suggesting potential applications across diverse interactive domains.
We present a system for visually transitioning a mixed reality (MR) user between two arbitrary realities (e.g., between two virtual worlds or between the real environment and a virtual world). The system uses artifici...
详细信息
ISBN:
(纸本)9798350374490;9798350374506
We present a system for visually transitioning a mixed reality (MR) user between two arbitrary realities (e.g., between two virtual worlds or between the real environment and a virtual world). The system uses artificial intelligence (AI) to generate a 360 degrees video that transforms the user's starting environment to another environment, passing through a liminal space that could help them relax between tasks or prepare them for the ending environment. The video can then be viewed on an MR headset.
In this paper, we explore the potential of virtual displays in supporting knowledge work, focusing on scenarios where conventional physical monitors fall short. Our research investigates the feasibility and productivi...
详细信息
ISBN:
(纸本)9798350374490;9798350374506
In this paper, we explore the potential of virtual displays in supporting knowledge work, focusing on scenarios where conventional physical monitors fall short. Our research investigates the feasibility and productivity costs of extending or replacing physical monitors with augmented and virtual reality displays. We present three compelling use cases, illustrating how virtual displays enhance flexibility, adaptability, and customization in remote and diverse work settings. Lessons learned from user studies highlight hardware and interface design challenges, emphasizing the need for larger resolution and field of view in head-worn displays (HWD). We conclude the paper with research opportunities and a call to address the evolving demands of knowledge work interfaces.
This paper explores user behavior in virtual reality (VR) environments, emphasizing the integration of idiosyncratic eye and head gaze for optimal user experience. Analyzing a gaze-based viewport control study dataset...
详细信息
ISBN:
(纸本)9798350374490;9798350374506
This paper explores user behavior in virtual reality (VR) environments, emphasizing the integration of idiosyncratic eye and head gaze for optimal user experience. Analyzing a gaze-based viewport control study dataset, we categorize users as head movers and non-head movers. In the study, participants were asked to align the viewport in VR so that a target is in the center. We tested three techniques: Controller Snap, Dwell Snap, and Gaze Pursuit, all using head and eye movement and showcasing particular behavioral patterns where some users prefer head movements, while others refrain from using head rotations for VR viewport control By that, our results point towards distinctive patterns of head- and non-head movers. The study highlights the need for personalized VR interfaces by considering these nuanced behavioral differences.
Dynamic field of view restriction is one effective way of mitigating cybersickness by modulating the amount of visual information during virtual navigation. However, in the presence of an interactive task for which vi...
详细信息
ISBN:
(纸本)9798350374490;9798350374506
Dynamic field of view restriction is one effective way of mitigating cybersickness by modulating the amount of visual information during virtual navigation. However, in the presence of an interactive task for which visibility is important, it can impede the task performance. This poster examined the efficiency of users, manually engaging the dynamic field of view restriction to control and mitigate cybersickness while performing interactive tasks. The comparative experiment has shown that the user-activated method significantly reduced cybersickness as much as the automatic method. However, it also achieved significantly higher task performance and usability despite the manual control.
暂无评论