Search results

Automated dust monitoring in workplaces helps provide timely alerts to over-exposed workers and effective mitigation measures for proactive dust control. However, the cluttered nature of construction sites poses a practical challenge to obtain enough high-quality images in the real world. The study aims to establish a framework that overcomes the challenges of lacking sufficient imagery data (“data-hungry problem”) for training computer vision algorithms to monitor construction dust.

This study develops a synthetic image generation method that incorporates virtual environments of construction dust for producing training samples. Three state-of-the-art object detection algorithms, including Faster-RCNN, you only look once (YOLO) and single shot detection (SSD), are trained using solely synthetic images. Finally, this research provides a comparative analysis of object detection algorithms for real-world dust monitoring regarding the accuracy and computational efficiency.

This study creates a construction dust emission (CDE) dataset consisting of 3,860 synthetic dust images as the training dataset and 1,015 real-world images as the testing dataset. The YOLO-v3 model achieves the best performance with a 0.93 F1 score and 31.44 fps among all three object detection models. The experimental results indicate that training dust detection algorithms with only synthetic images can achieve acceptable performance on real-world images.

This study provides insights into two questions: (1) how synthetic images could help train dust detection models to overcome data-hungry problems and (2) how well state-of-the-art deep learning algorithms can detect nonrigid construction dust.

Having been a promising visualization tool since 1950s, ironically, virtual reality is not widely used in the architectural design and evaluation process due to several constrains, such as the high cost of equipments and advanced programming skills required. This paper described the collaboration between design computing courses and architecture design studios that have been taught at Savannah College of Art and Design (SCAD) in 2004 and 2005. These courses explored several practical methods to integrate Low Cost Virtual Reality Aided Design (LC-VRAD) in the architectural design process. As a summary of the collaboration, this paper refers to three main aspects: (1) How to use game engine to design an affordable VR system in the ordinary studio environment. (2) How to integrate VR, into the design process, not only as a visualization tool, but also as a design instrument. (3) How to evaluate different methods of representing architectural models based on the efficiency of workflow, rendering quality and users' feedback.

Support by the Game and Interactive Design Department at SCAD, students in the School of Building Arts implemented two Low Cost VRAD methods in various design phases, starting from site analysis, schematic design, design development to the final presentation. Two popular game engines, Epic Game's Unreal engine and Director MX's Shockwave engine, were introduced to students to visualize their project in real-time. We discussed computer-aided design theories including the application of VR, as well as digital computing and human computer interaction. At the end of each quarter, feedbacks from students and faculties were collected and analyzed. These methods were revised and improved consistently across 2004 and 2005 academic year.

In this study, a new methodology to evaluate the performance of physics simulation engines (PSEs) when used in haptic virtual assembly applications is proposed. This methodology can be used to assess the performance of any physics engine. To prove the feasibility of the proposed methodology, two-third party PSEs – Bullet and PhysXtm – were evaluated. The paper aims to discuss these issues.

Eight assembly tests comprising variable geometric and dynamic complexity were conducted. The strengths and weaknesses of each simulation engine for haptic virtual assembly were identified by measuring different parameters such as task completion time, influence of weight perception and force feedback.

The proposed tests have led to the development of a standard methodology by which physics engines can be compared and evaluated. The results have shown that when the assembly comprises complex shapes, Bullet has better performance than PhysX. It was also observed that the assembly time is directly affected by the weight of virtual objects.

A more comprehensive study must be carried out in order to evaluate and compare the performance of more PSEs. The influence of collision shape representation algorithms on the performance of haptic assembly must be considered in future analysis.

The performance of PSEs in haptic-enabled VR applications had been remained as an unknown issue. The main parameters of physics engines that affect the haptic virtual assembly process have been identified. All the tests performed in this study were carried out with the haptic rendering loop active and the objects manipulated through the haptic device.

The purpose of this paper is to give a comprehensive survey on the physics-based virtual assembly (PBVA) technology in a novel perspective, to analyze current drawbacks and propose several promising future directions.

To provide a deep insight of PBVA, a discussion of the developing context of PBVA and a comparison against constraint-based virtual assembly (CBVA) is put forward. The core elements and general structure are analyzed based on typical PBVA systems. Some common key issues as well as common drawbacks are discussed, based on which the research trend and several promising future directions are proposed.

Special attention is paid to new research progresses and new ideas concerning recent development as well as new typical systems of the technology. Advantages of PBVA over CBVA are investigated. Based on the analysis of typical PBVA systems and the evolution of PBVA, the core elements of the technology and the general structure of its implementation are identified. Then, current PBVA systems are summarized and classified. After that, key issues in the technology and current drawbacks are explored in detail. Finally, promising future directions are given, including both the further perfecting of the technology and the combination with other technologies.

The PBVA technology is put into a detailed review and analysis in a novel way, providing a better insight of both the theory and the implementation of the technology.

Realistic force sensation can help operators better feel and manipulate parts for virtual assembly (VA). Moreover, for VA of mechanical parts, it is necessary to consider their tolerance levels so as to apply proper assembly forces. Out of the three common assembly fit types, the type of clearance fit is the focus of virtual manual assembly, as parts with such fit type require precise force feedback to assist users’ assembly operations.

This study proposes a novel force rendering model for VA of mechanical parts with clearance fits. By decomposing an actual assembly operation into three consecutive states, the corresponding forces are formulated.

A prototype system is designed and developed to implement the model, and comparative case studies are conducted to investigate the users’ performance with the other three common approaches, namely, a typical WIMP (window-icon-menu-pointer) interface with CAD software, a physics simulation with collision detection and the approach that combines physics simulation and geometric constraints restriction. The results have shown that the proposed model is more realistic by providing continuous and realistic force feedback to the users.

The users’ feeling of immersion and their operational efficiency are greatly enhanced with the force sensation provided.

Haptics is an emerging technology that allows touch‐enabled interaction with virtual objects. Analogous to the use of computer graphics for rendering of a three‐dimensional (3D) scene to give the user a visual description of the scene, it is possible to use computer haptics to let the user touch objects in the 3D scene. This is normally accomplished by having the haptics engine sending either force vectors or positional information to a haptics device, a robotic arm, that the user manipulates. The purpose of this paper is to give an overview of this technology, describe haptic devices and haptic application programming interfaces. We will also illustrate the use of haptics technology by describing a few industrial and medical applications.

Theatre Engine is a three-phase project exploring the integration of mobile devices such as smart phones and tablet computers as user interaction mechanisms for live theatrical performances. This paper aims to describe the technical details for the first performance in the series: transitions, including system design, user interface details, and lessons learned.

The Theatre Engine project consists of a network of mobile Android-based devices that communicate with a server managing a live theatrical performance. Additional systems provide projected graphics within the performance space, dynamic instructions to live musicians, and control of theatrical equipment including robotic lighting. The system is under the live control of a stage managers and show runner. The first phase produced the performance transitions, where live and virtual dancers interacted with audience members.

The user activities are described in the context of manipulations and the effects of those manipulations and a classification of the roles of users as active participants in the performance is proposed. The concept of blurring the distinction between real and virtual is explored.

Technological advances make it possible for the audience to be an active participant in a performance, converting a passive theatrical experience to an active and immersive experience. Theatre Engine is a new approach to the integration of mobile devices into the theatre and a test-bed for new user interface ideas.

The purpose of this paper is to present the investigation on contemporary applications of game design in architectural visualization, urban environmental planning and the results to address the problems of traditional methods specifically in terms of interactivity and visualization. The authors present the prototype that incorporates information modeling and virtual reality (VR) into interactive architectural visualization.

The proposed system supports a virtual walkthrough process that allows users to navigate in an architectural space of their imagination and design. The immersion in the design of living space or environment has been made possible through the inclusion of VR through the use of WebVR technology to deploy the design to be experienced.

This study investigates and establishes a framework that explores the intricacies of 3 imensional (3D) Architectural Visualization in a real-time engine by designing a visual experience with a better level of user interaction and then deploying it through VR.

To the best of the authors’ knowledge, this is one of the few experimental frameworks which perfectly integrates VR for visualization in digital home design environments. The prospective applications of this framework are in several fields of construction innovation, architectural design, 3D modeling, virtual and augmented reality.

An emerging, globally Web‐based Chinese language virtual library offers unprecedented content availability and user accessibility. Virtual and physical libraries are defined, and compared, in terms of bibliographical searching. Then, the size of the Chinese language virtual library is estimated and its future development is predicted. The quantitative analysis concentrates on the content of this virtual library through examining the subject directories of more than 100 of its search engines. The findings show that 20 percent of the subject categories provide 80 percent of online access activities. Many subject categories characterise values of the Chinese‐speaking world. Discussion continues with structural and functional analyses of Chinese language search engines that support this virtual library. Further, an analytic overview is presented of Chinese language applications that render the necessary and sufficient computational linguistic conditions for utilising Chinese language online resources.

The purpose of this paper is to discuss the use of stereoscopic virtual technology in textile and fashion studies in particular to the area of chemical experiment. The development of a designed virtual platform, called Stereoscopic Chemical Laboratory (SCL), is introduced.

To implement the suggested educational approaches of SCL, a set of teaching and learning materials with emphasis on the application methods was revised from the existing subjects. The architecture of SCL includes building of virtual objects with the Autodesk software Maya and designing of interactivity by using Unity, a game engine system. Prototype version of the SCL has been passed to selected academic colleagues and students for further evaluation and application feedback.

Textile students can conduct laboratory experiments associated with coloration and finishing of textile technologies in a stereoscopic 3D and multisensory laboratory, and hence enhance their learning experience. With the use of SCL, students can learn relevant experiment tools, experimental processes, procedures, and safety and health precautions.

There is very limited educational or training approach in applying stereoscopic virtual reality in teaching activities. In the area of textile experiment, the authors could say it is virtual and does not exist in current research domains.