VR Space Research Articles

A novel AR remote collaborative platform for sharing 2.5D gestures and gaze

Although sharing gestures and gaze can improve AR remote collaboration, most current systems only enable collaborators to share 2D or 3D gestures, and the unimodal HCI interface remains dominant. To address this problem, we describe a novel remote collaborative platform based on 2.5D gestures and gaze (2.5DGG), which supports an expert who collaborates with a worker (e.g., during assembly or training tasks). We investigate the impact of sharing the remote site’s 2.5DGG using spatial AR (SAR) remote collaboration in manufacturing. Compared to other systems, there is a key advantage that it can provide more natural and intuitive multimodal interaction based on 2.5DGG. We track the remote experts’ gestures and eye gaze using Leap Motion and aGlass, respectively, in a VR space displaying the live video stream of the local physical workspace and visualize them onto the local work scenario by a projector. The results of an exploratory user study demonstrate that 2.5DGG has a clear difference in performance time and collaborative experience, and it is better than the traditional one.

Development of a platform part wearable force feedback device that can present the sensation of dropping in a VR space

Development of a platform part wearable force feedback device that can present the sensation of dropping in a VR space

A study on self-body recognition for a four-armed avatar simultaneously controlled by two people in a VR space

A study on self-body recognition for a four-armed avatar simultaneously controlled by two people in a VR space

Development of a wearable lower limb force feedback device that can move freely in VR space

Development of a wearable lower limb force feedback device that can move freely in VR space

An Analysis of Personalized Learning Opportunities in 3D VR

Due to its constantly developing technological background, VR and AR technology has been gaining increasing popularity not just in industry or business but in education as well. Research in the field of Cognitive Infocommunications (CogInfoCom) shows that using existing digital technologies, online collaboration and cooperation technologies in 3D VR supports cognitive processes, including the finding, processing, memorization and recalling of information. 3D VR environments are also capable of providing users with a much higher level of comprehension when it comes to sharing and interpreting digital workflows. The paper presents a study carried out with the participation of 90 students. The aim of this study is to investigate how the application of 3D VR platforms as personalized educational environments can also increase VR learning efficiency. Besides considering participants’ test performance, metrics such as results on visual, auditory and reading-based learning tests for information acquisition, as well as responses on Kolb’s learning styles questionnaires are taken into consideration. The participants’ learning styles, information acquisition habits were also observed, allowing us to create and offer a variety of learning pathways based on a variety of content types in the 3D VR environment. The students within the study were divided into two groups: a test group receiving personalized training in the MaxWhere 3D VR classroom, and a control group that studied in a general MaxWhere 3D VR space. This research applies both quantitative and qualitative methods to report findings. The goal was to create adaptive learning environments capable of deriving models of learners and providing personalized learning experiences. We studied the correlation between effectiveness of the tasks and Kolb’s learning styles. The study shows the major importance of choosing the optimal task type regarding each Kolb learning style and personalized learning environment. The MaxWhere 3D spaces show a high potential for personalizing VR education. The non-intrusive guiding capabilities of VR environments and of the educational content integrated in the 3D VR spaces were very successful, because the students were able to score 20 percent higher on the tests after studying in VR than after using traditional educational tools. Students also performed the same tasks with 8-10 percent faster response times.

Visuomotor control of intermittent circular tracking movements with visually guided orbits in 3D VR environment.

The analysis of visually guided tracking movements is important to the understanding of imitation exercises and movements carried out using the human visuomotor control system. In this study, we analyzed the characteristics of visuomotor control in the intermittent performance of circular tracking movements by applying a system that can differentiate between the conditions of invisible and visible orbits and visible and invisible target phases implemented in a 3D VR space. By applying visuomotor control based on velocity control, our study participants were able to track objects with visible orbits with a precision of approximately 1.25 times greater than they could track objects with invisible orbits. We confirmed that position information is an important parameter related to intermittent motion at low speeds (below 0.5 Hz) and that tracked target velocity information could be obtained more precisely than position information at speeds above 0.5 Hz. Our results revealed that the feedforward (FF) control corresponding to velocity was delayed under the visible-orbit condition at speeds over 0.5 Hz, suggesting that, in carrying out imitation exercises and movements, the use of visually presented 3D guides can interfere with exercise learning and, therefore, that the effects of their use should be carefully considered.

Templated Text Synthesis for Expert-Guided Multi-Label Extraction from Radiology Reports

Training medical image analysis models traditionally requires large amounts of expertly annotated imaging data which is time-consuming and expensive to obtain. One solution is to automatically extract scan-level labels from radiology reports. Previously, we showed that, by extending BERT with a per-label attention mechanism, we can train a single model to perform automatic extraction of many labels in parallel. However, if we rely on pure data-driven learning, the model sometimes fails to learn critical features or learns the correct answer via simplistic heuristics (e.g., that “likely” indicates positivity), and thus fails to generalise to rarer cases which have not been learned or where the heuristics break down (e.g., “likely represents prominent VR space or lacunar infarct” which indicates uncertainty over two differential diagnoses). In this work, we propose template creation for data synthesis, which enables us to inject expert knowledge about unseen entities from medical ontologies, and to teach the model rules on how to label difficult cases, by producing relevant training examples. Using this technique alongside domain-specific pre-training for our underlying BERT architecture i.e., PubMedBERT, we improve F1 micro from 0.903 to 0.939 and F1 macro from 0.512 to 0.737 on an independent test set for 33 labels in head CT reports for stroke patients. Our methodology offers a practical way to combine domain knowledge with machine learning for text classification tasks.

Evaluation of Interaction with Musculo-skeletal Avatar in VR Space

Evaluation of Interaction with Musculo-skeletal Avatar in VR Space

Factors Affecting Dynamic Moving Detectability in VR Space

Factors Affecting Dynamic Moving Detectability in VR Space

The effect of using a controller to adjust gaze pointing in VR space

The effect of using a controller to adjust gaze pointing in VR space

VR공간에서의 플로우 형성과정 연구

VR공간에서의 플로우 형성과정 연구

A Tactile Sense Centered Virtual Reality game by using Biometric feedback

We construct a Virtual Reality, VR system centered on the tactile sensation. Tactile sensation has a more significant effect on sensibility than audiovisual. Therefore, if tactile sensation and audiovisual sense are combined, we can create “virtual presence” in the VR space. In this system, a user’s biological information is linked to the audiovisual data presented. A player plays with virtual objects through physical body movements. And the system feeds-back biometric information and deliver the change of biometric data through the sensory presentation to the user.

Deep-Learning-Based Stress-Ratio Prediction Model Using Virtual Reality with Electroencephalography Data

The Reich Chancellery, built by Albert Speer, was designed with an overwhelming ambience to represent the worldview of Hitler. The interior of the Reich Chancellery comprised high-ceiling and low-ceiling spaces. In this study, the change in a person’s emotions according to the ceiling height while moving was examined through brain wave experiments to understand the stress index for each building space. The Reich Chancellery was recreated through VR, and brain wave data collected per space were processed through a first and second analysis. In the first analysis, beta wave changes related to the stress index were calculated, and the space with the highest fluctuation was analyzed. In the second analysis, the correlation between 10 different types of brain waves and waveforms was analyzed; deep-learning algorithms were used to verify the accuracy and analyze spaces with a high stress index. Subsequently, a deep-learning platform for calculating such a value was developed. The results showed that the change in stress index scores was the highest when entering from the Mosaic Hall (15 m floor height) to the Führerbunker (3 m floor height), which had the largest floor height difference. Accordingly, a stress-ratio prediction model for selecting a space with a high stress level was established by monitoring the architectural space based on brain wave information in a VR space. In the architectural design process, the ratio can be used to reflect user sensibility in the design and improve the efficiency of the design process.

Analysis of Three-Dimensional Circular Tracking Movements Based on Temporo-Spatial Parameters in Polar Coordinates

Motor control characteristics of the human visuomotor control system need to be analyzed in the three-dimensional (3D) space to study and imitate human movements. In this paper, we examined circular tracking movements on two planes in 3D space from a motor control perspective based on three temporospatial parameters in polar coordinates. Sixteen healthy human subjects participated in this study and performed circular target tracking movements rotating at 0.125, 0.25, 0.5, and 0.75 Hz in the frontal or sagittal planes in three-dimensional space. The results showed that two temporal parameter errors on each plane were proportional to the change in the target velocity. Furthermore, frontal plane circular tracking errors without depth for a spatial parameter were lower than those for sagittal plane circular tracking with depth. The experimental protocol and data analysis allowed us to analyze the motor control characteristics temporospatially for circular tracking movement with various depths and speeds in the 3D VR space.

Dementia Simulator by Manipulation of Out-of-Sight VR Space

Dementia Simulator by Manipulation of Out-of-Sight VR Space

Speech Support System Using a Noise Environment Including Multiple Non-participants in VR Space

Speech Support System Using a Noise Environment Including Multiple Non-participants in VR Space

VRCI and CPHS

Abstract Virtual Reality of Complete Immersion (VRCI) is the ultimate form of VR, which means that all human beings can immerse in it permanently and carry out any activities that we accomplish in physical world or meet any need to be met for living. VRCI is now a theoretical assumption, but the technologies it based on are seamless continuous with our existing technologies. VRCI can be regarded as a special CPHS, meanwhile very different from a typical one, particularity in VR space compared with physical space, cyber space and social space. Compared with the design idea of existing CPS, CPHS or CPSS, VRCI provides human beings a totally edible living space which is separated from the physical space, and even will change the form of human civilization in the future.

A flexible technique to select objects via convolutional neural network in VR space

Most studies on the selection techniques of projection-based VR systems are dependent on users wearing complex or expensive input devices, however there are lack of more convenient selection techniques. In this paper, we propose a flexible 3D selection technique in a large display projection-based virtual environment. Herein, we present a body tracking method using convolutional neural network (CNN) to estimate 3D skeletons of multi-users, and propose a region-based selection method to effectively select virtual objects using only the tracked fingertips of multi-users. Additionally, a multi-user merge method is introduced to enable users actions and perception to realign when multiple users observe a single stereoscopic display. By comparing with state-of-the-art CNN-based pose estimation methods, the proposed CNN-based body tracking method enables considerable estimation accuracy with the guarantee of real-time performance. In addition, we evaluate our selection technique against three prevalent selection techniques and test the performance of our selection technique in a multi-user scenario. The results show that our selection technique significantly increases the efficiency and effectiveness, and is of comparable stability to support multi-user interaction.

Usability Evaluation Model for Locomotion Technology in VR Space

Usability Evaluation Model for Locomotion Technology in VR Space

The influence of time perception by VR space

The influence of time perception by VR space