Deep learning-based object detection in augmented reality: A systematic review

Ghasemi, Yalda; Jeong, Heejin; Choi, Sung Ho; Park, Kyeong-Beom; Lee, Jae Yeol

doi:10.1016/j.compind.2022.103661

Cited by 88 publications

(29 citation statements)

References 77 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The emergence of markerless AR systems allows objects to be identified and tracked by capturing and processing images of those objects. The application of deep learning for object detection and identification using markerless AR technology is reported to be fast as well as accurate [ 19 ]. However, the limitation of such approaches is the associated computational burden that might occur.…”

Section: Augmented Reality Technologiesmentioning

confidence: 99%

“…There are many types of wearable AR devices currently available, among which the most popular include Microsoft HoloLens™, Google Glass™, Epson Moverio™, Vuzix Blade™, Magic Leap™, and GlassUp F4 Smart Glasses™. Such wearable AR devices, also known as optical see-through (OST) heads-up displays, are considered to be the most advanced [ 19 ]. These wearables have the advantage of not impeding the user’s vision and at the same time allowing better interactions with the user by facilitating hands-free use.…”

Section: Augmented Reality Technologiesmentioning

confidence: 99%

See 1 more Smart Citation

Trends of Augmented Reality for Agri-Food Applications

Xie

Chai²,

O’Sullivan³

et al. 2022

Sensors

View full text Add to dashboard Cite

Recent years have witnessed an increasing interest in deploying state-of-the-art augmented reality (AR) head-mounted displays (HMDs) for agri-food applications. The benefits of AR HMDs to agri-food industry stakeholders (e.g., food suppliers, retail/food service) have received growing attention and recognition. AR HMDs enable users to make healthier dietary choices, experience novel changes in their perception of taste, enhance the cooking and food shopping experience, improve productivity at work and enhance the implementation of precision farming. Therefore, although development costs are still high, the case for integration of AR in food chains appears to be compelling. This review will present the most recent developments of AR HMDs for agri-food relevant applications. The summarized applications can be clustered into different themes: (1) dietary and food nutrition assessment; (2) food sensory science; (3) changing the eating environment; (4) retail food chain applications; (5) enhancing the cooking experience; (6) food-related training and learning; and (7) food production and precision farming. Limitations of current practices will be highlighted, along with some proposed applications.

show abstract

Section: Augmented Reality Technologiesmentioning

confidence: 99%

Section: Augmented Reality Technologiesmentioning

confidence: 99%

Trends of Augmented Reality for Agri-Food Applications

Xie

Chai²,

O’Sullivan³

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…The use of DL in AR is still relatively new. Several previous studies have proposed the use of AR, such as [8], [10] (DNN), [11] (LSTM), and [12][13][14][15] (CNN). One study [16] used a deep neural network (DNN) for an intelligent municipality AR service system in the fields of information dissemination and tourism.…”

Section: Related Workmentioning

confidence: 99%

Object Tracking in Augmented Reality: Enhancement Using Convolutional Neural Networks

Nurhadi

Stiawan

Idris

et al. 2022

IJEEI

View full text Add to dashboard Cite

Augmented reality (AR) has been applied in maintenance, simulation, remote assistance, and other fields. One of the issues arising from these applications is how objects can be placed in physical environments using an AR system. Object placement consists of two processes: object detection and segmentation. Due to the importance of placement, in this paper, we propose using deep learning to address issues with the placement of objects through detection and segmentation in AR. Deep learning can help complete tasks by providing correct information about environmental changes in real-time situations. The problem is that it is rarely used in AR, which suggests a combination of deep learning-based object detection and instance segmentation with wearable AR technology to improve performance on complex tasks. In our work, we propose to address this problem by applying a convolutional neural network to facilitate the detection and segmentation of objects in real environments. To measure AR performance, we examined detection accuracy in environments with different intensities. The results of the experiment demonstrate satisfactory performance, reaching 98% for segmentation and accurate detection.

show abstract

“…In the case of wave-transparent GFRP honeycomb structures, such as radomes, water ingress will severely degrade wave-transparent performance. Therefore, the detection and classification of liquid ingress in honeycomb structures becomes an essential step in the manufacturing process and during service [ 1 , 2 , 3 , 4 ].…”

Section: Introductionmentioning

confidence: 99%

Classification of Liquid Ingress in GFRP Honeycomb Based on One-Dimension Sequential Model Using THz-TDS

Huo

et al. 2023

Sensors

View full text Add to dashboard Cite

Honeycomb structure composites are taking an increasing proportion in aircraft manufacturing because of their high strength-to-weight ratio, good fatigue resistance, and low manufacturing cost. However, the hollow structure is very prone to liquid ingress. Here, we report a fast and automatic classification approach for water, alcohol, and oil filled in glass fiber reinforced polymer (GFRP) honeycomb structures through terahertz time-domain spectroscopy (THz-TDS). We propose an improved one-dimensional convolutional neural network (1D-CNN) model, and compared it with long short-term memory (LSTM) and ordinary 1D-CNN models, which are classification networks based on one dimension sequenced signals. The automated liquid classification results show that the LSTM model has the best performance for the time-domain signals, while the improved 1D-CNN model performed best for the frequency-domain signals.

show abstract

Deep learning-based object detection in augmented reality: A systematic review

Cited by 88 publications

References 77 publications

Trends of Augmented Reality for Agri-Food Applications

Trends of Augmented Reality for Agri-Food Applications

Object Tracking in Augmented Reality: Enhancement Using Convolutional Neural Networks

Classification of Liquid Ingress in GFRP Honeycomb Based on One-Dimension Sequential Model Using THz-TDS

Contact Info

Product

Resources

About