Applications of Smart Glasses in Applied Sciences: A Systematic Review

Kim, Dawon; Choi, Yosoon

doi:10.3390/app11114956

Cited by 46 publications

(25 citation statements)

References 66 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although the smartphone is one type of device that can host an augmented reality application by overlaying information on the display incorporating a built-in camera, deploying the solution to a body-worn device such as smart glasses would be more intuitive because such a device is able to project information directly through an optical head-mounted display (OHMD). Kim and Choi [ 33 ] surveyed 57 academic papers on the applications of smart glasses and stated that smart glasses are most often used in the healthcare domain ( n = 21, 37%). Evaluating the performance of caregivers with and without augmented reality would be an interesting prospective study.…”

Section: Discussionmentioning

confidence: 99%

Rapid Convolutional Neural Networks for Gram-Stained Image Classification at Inference Time on Mobile Devices: Empirical Study from Transfer Learning to Optimization

et al. 2022

View full text Add to dashboard Cite

Despite the emergence of mobile health and the success of deep learning (DL), deploying production-ready DL models to resource-limited devices remains challenging. Especially, during inference time, the speed of DL models becomes relevant. We aimed to accelerate inference time for Gram-stained analysis, which is a tedious and manual task involving microorganism detection on whole slide images. Three DL models were optimized in three steps: transfer learning, pruning and quantization and then evaluated on two Android smartphones. Most convolutional layers (≥80%) had to be retrained for adaptation to the Gram-stained classification task. The combination of pruning and quantization demonstrated its utility to reduce the model size and inference time without compromising model quality. Pruning mainly contributed to model size reduction by 15×, while quantization reduced inference time by 3× and decreased model size by 4×. The combination of two reduced the baseline model by an overall factor of 46×. Optimized models were smaller than 6 MB and were able to process one image in <0.6 s on a Galaxy S10. Our findings demonstrate that methods for model compression are highly relevant for the successful deployment of DL solutions to resource-limited devices.

show abstract

Section: Discussionmentioning

confidence: 99%

Rapid Convolutional Neural Networks for Gram-Stained Image Classification at Inference Time on Mobile Devices: Empirical Study from Transfer Learning to Optimization

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Architecture and engineering design [17,21,23,[29][30][31][32][33][34][35][36][37][38][39][40][41][42][43][44][45][46] Building construction [17,[20][21][22][23][24][25]27,[29][30][31][47][48][49][50][51][52][53][54][55][56] Building operation and management [17,18,21,23,26,28,54, Selected articles concerning XR application in building operation and management were further divided into VR and AR/MR categories, each articulated in different domains according to their possible use in building facility management as shown in Table…”

Section: Xr Application Field Selected Articlesmentioning

confidence: 99%

“…Overall, AR can be implemented according to four methods: optical see-through, video see-through, eye multiplexed, and projection based. The former two are the most widely adopted in AR headsets available on the market [56]. In optical see-through systems, AR is achieved by superimposing virtual images over the direct view of the real world, commonly by projecting CG content through half mirrors or prisms.…”

Section: Cg Content Overlay Mode Descriptionmentioning

confidence: 99%

Extended Reality for Smart Building Operation and Maintenance: A Review

Casini

2022

Energies

View full text Add to dashboard Cite

The operation and maintenance (O&M) of buildings and infrastructure represent a strategic activity to ensure they perform as expected over time and to reduce energy consumption and maintenance costs at the urban and building scale. With the increasing diffusion of BIM, IoT devices, and AI, the future of O&M is represented by digital twin technology. To effectively take advantage of this digital revolution, thus enabling data-driven energy control, proactive maintenance, and predictive daily operations, it is vital that smart building management exploits the opportunities offered by the extended reality (XR) technologies. Nevertheless, in consideration of the novelty of XR in the AECO sector and its rapid and ongoing evolution, knowledge of the specific possibilities and the methods of integration into the building process workflow is still piecemeal and sparse. With the goal to bridge this gap, the article presents a thorough review of virtual reality (VR), augmented reality (AR), and mixed reality (MR) technologies and applications for smart building operation and maintenance. After defining VR, AR, and MR, the article provides a detailed review that analyzes, categorizes, and summarizes state-of-the-art XR technologies and their possible applications for building O&M along with their relative advantages and disadvantages. The article concludes that the application of XR in building and city management is showing promising results in enhancing human performance in technical O&M tasks, in understanding and controlling the energy efficiency, comfort, and safety of building and infrastructures, and in supporting strategic decision making for the future smart city.

show abstract

“…Systematic literature review (SLR) is a method for identifying, evaluating, and synthesizing the work produced by researchers [18], whereas bibliometric analysis is a statistical evaluation of research documents [19]. Number of articles, collaboration networks, data inspection, and research methodologies are just a few examples of information gathered using SLR and bibliometric analysis in fields such as material selection processes [20], polymers manufacturing [21], biomaterials for implants [22], and smart glasses applications [23]. Here, through SLR and bibliometric analysis, we provide answers to the following questions on MGs studied by means of MD simulations: When did this research field grow?…”

Section: Introductionmentioning

confidence: 99%

Research on metallic glasses at the atomic scale: a systematic review

Amigó

Cortés²,

Valencia

2022

SN Appl. Sci.

View full text Add to dashboard Cite

Metallic glasses (MGs) have been long investigated in material science to understand the origin of their remarkable properties. With the help of computational simulations, researchers have delved into structure-property relationships, leading to a large number of reports. To quantify the available literature, we employed systematic review and bibliometric analysis on studies related to MGs and classical molecular dynamics simulations from 2000 to 2021. It was found that the total number of articles has increased remarkably, with China and the USA producing more than half of the reports. However, high-impact articles were mainly conducted in the latter. Collaboration networks revealed that top contributor authors are strongly connected with other researchers, which emphasizes the relevance of scientific cooperation. In regard to the evolution of research topics, according to article keywords, plastic behavior has been a recurrent subject since the early 2000s. Nevertheless, the traditional approach of studying monolithic MGs at the short-range order evolved to complex composites with characterizations at the medium-range order, including topics such as nanoglasses, amorphous/crystalline nanolaminates, rejuvenation, among others. As a whole, these findings provide researchers with an overview of past and current trends of research areas, as well as some of the leading authors, productivity statistics, and collaboration networks.

show abstract

Applications of Smart Glasses in Applied Sciences: A Systematic Review

Cited by 46 publications

References 66 publications

Rapid Convolutional Neural Networks for Gram-Stained Image Classification at Inference Time on Mobile Devices: Empirical Study from Transfer Learning to Optimization

Rapid Convolutional Neural Networks for Gram-Stained Image Classification at Inference Time on Mobile Devices: Empirical Study from Transfer Learning to Optimization

Extended Reality for Smart Building Operation and Maintenance: A Review

Research on metallic glasses at the atomic scale: a systematic review

Contact Info

Product

Resources

About