ICRICS: iterative compensation recovery for image compressive sensing

Li, Honggui; Trocan, Maria; Sawan, Mohamad; Galayko, Dimitri

doi:10.1007/s11760-023-02516-z

Cited by 2 publications

(1 citation statement)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, the digital zoom algorithm used in mobile cameras and the image enhancement techniques used in digital devices. Furthermore, this core technology can be applied to a wide range of Computer Vision tasks, which leads to improvements in various Vision tasks, such as object detection [2], [3], medical imaging [4], [5], security and surveillance imaging [6], [7], face recognition [8], [9].…”

Section: Introductionmentioning

confidence: 99%

SRFormer: Efficient Yet Powerful Transformer Network for Single Image Super Resolution

Mehri,

Behjati,

Carpio

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Recent breakthroughs in single image super resolution have investigated the potential of deep Convolutional Neural Networks (CNNs) to improve performance. However, CNNs based models suffer from their limited fields and their inability to adapt to the input content. Recently, Transformer based models were presented, which demonstrated major performance gains in Natural Language Processing and Vision tasks while mitigating the drawbacks of CNNs. Nevertheless, Transformer computational complexity can increase quadratically for high-resolution images, and the fact that it ignores the original structures of the image by converting them to the 1D structure can make it problematic to capture the local context information and adapt it for real-time applications. In this paper, we present, SRFormer, an efficient yet powerful Transformer-based architecture, by making several key designs in the building of Transformer blocks and Transformer layers that allow us to consider the original structure of the image (i.e., 2D structure) while capturing both local and global dependencies without raising computational demands or memory consumption. We also present a Gated Multi-Layer Perceptron (MLP) Feature Fusion module to aggregate the features of different stages of Transformer blocks by focusing on inter-spatial relationships while adding minor computational costs to the network. We have conducted extensive experiments on several super-resolution benchmark datasets to evaluate our approach. SRFormer demonstrates superior performance compared to state-of-the-art methods from both Transformer and Convolutional networks, with an improvement margin of 0.1 ∼ 0.53dB. Furthermore, while SRFormer has almost the same model size, it outperforms SwinIR by 0.47% and inference time by half the time of SwinIR. The code will be available on GitHub.

show abstract

Section: Introductionmentioning

confidence: 99%

SRFormer: Efficient Yet Powerful Transformer Network for Single Image Super Resolution

Mehri,

Behjati,

Carpio

et al. 2023

IEEE Access

View full text Add to dashboard Cite

show abstract

An Empirical Survey on Wireless Inductive Power Pad and Resonant Magnetic Field Coupling for In-Motion EV Charging System

et al. 2023

View full text Add to dashboard Cite

EVs are the recent emerging automotive technology in the transportation sector to reduce the CO2 emission from the internal combustion engine. The issues in EVs technology development are battery tube capacity, heavy-size batteries, fast charging, and safe charging infrastructure. The dynamic wireless charging technology shows a suitable alternative to address the charging system-related issues in EV. However, a limited number of review studies are conducted to specifically address the wireless charging pad design challenges. The wireless inductive power pad and magnetic coupling circuit design are the main factors to decide the performance of the DWPT system. This review analyzes the current developments and challenges associated with wireless charging pad design. Further, this study investigates the potential parameters which improve the performance of a DWPT system to increase the distance traveled (mileage). First, this paper discusses WRIPT technology for DWPT EV charging application, and several parameters affecting the PTE are examined. Also, the aids factors considered for designing the DWPT power pad and different magnetic resonance coupling topologies are presented. In addition, the performance evaluation of the WRIPT power pad, with in-motion testing from the major findings in earlier studies is discussed. Finally, the challenges and opportunities of the WRIPT power pad for in-motion EV charging applications are also addressed. The current state of the art of DWPT and its future directions to make DWPT EV charging systems a fullfledged method are highlighted.

show abstract

ICRICS: iterative compensation recovery for image compressive sensing

Cited by 2 publications

References 63 publications

SRFormer: Efficient Yet Powerful Transformer Network for Single Image Super Resolution

SRFormer: Efficient Yet Powerful Transformer Network for Single Image Super Resolution

An Empirical Survey on Wireless Inductive Power Pad and Resonant Magnetic Field Coupling for In-Motion EV Charging System

Contact Info

Product

Resources

About