Image Super-Resolution Based on Generative Adversarial Networks: A Brief Review

Single image super resolution (SISR) is an important research content in the field of computer vision and image processing. With the rapid development of deep neural networks, different image super-resolution models have emerged. Compared to some traditional SISR methods, deep learning-based methods can complete the superresolution tasks through a single image. In addition, compared with the SISR methods using traditional convolutional neural networks, SISR based on generative adversarial networks (GAN) has achieved the most advanced visual performance. In this review, we first explore the challenges faced by SISR and introduce some common datasets and evaluation metrics. Then, we review the improved network structures and loss functions of GAN-based perceptual SISR. Subsequently, the advantages and disadvantages of different networks are analyzed by multiple comparative experiments. Finally, we summarize the paper and look forward to the future development trends of GAN-based perceptual SISR.     

Compiler compatible 5.66 Mb/mm2 8T 1R1W register file in 14 nm FinFET technology

1-read/1-write (1R1W) register file (RF) is a popular memory configuration in modern feature rich SoCs requiring significant amount of embedded memory. A memory compiler is constructed using the 8T RF bitcell spanning a range of instances from 32 b to 72 Kb. An 8T low-leakage bitcell of 0.106 μm² is used in a 14 nm FinFET technology with a 70 nm contacted gate pitch for high-density (HD) two-port (TP) RF memory compiler which achieves 5.66 Mb/mm² array density for a 72 Kb array which is the highest reported density in 14 nm FinFET technology. The density improvement is achieved by using techniques such as leaf-cell optimization (eliminating transistors), better architectural planning, top level connectivity through leaf-cell abutment and minimizing the number of unique leaf-cells. These techniques are fully compatible with memory compiler usage over the required span. Leakage power is minimized by using power-switches without degrading the density mentioned above. Self-induced supply voltage collapse technique is applied for write and a four stack static keeper is used for read Vmin improvement. Fabricated test chips using 14 nm process have demonstrated 2.33 GHz performance at 1.1 V/25 °C operation. Overall V_min of 550 mV is achieved with this design at 25 °C. The inbuilt power-switch improves leakage power by 12x in simulation. Approximately 8% die area of a leading 14 nm SoC in commercialization is occupied by these compiled RF instances.     

高效AVS音频上下文位平面编码器设计

研究了AVS音频编码标准中上下文位平面编码算法,分析了编码算法的特点。通过对上下文平面编码算法的优化,提出了一种适合ASIC实现的硬件结构,通过合理设计二级流水线,能近似达到每时钟1比特矢量的编码速率,在保证编码速度前提下大幅降低了硬件资源。     

High birefringence photonic bandgap fiber with elliptical air holes

In this paper, we proposed a high birefringence photonic bandgap fiber (PBGF) with elliptical air holes in cladding and circular air hole in core. The forbidden gaps of the honeycomb structure are calculated when the circular air holes are deformed to the elliptical shape. And the birefringence of the PBGFs is investigated by using a full-vector finite element method (FEM) and the numerical results are presented.     

Investigation of performance variations due to the amplitude of group-delay ripple in chirped fiber Bragg gratings

We measure the system impacts due to the amplitude of group-delay (GD) ripple in single and cascaded chirped fiber Bragg gratings (CFBGs). Signals with smaller pulse width result in smaller performance variation at the same data rate. A 65-ps peak-to-peak GD ripple induces 0.9, 1.7, and 2.7 dB maximum penalties for 10, 20, and 40-Gb/s, respectively. We also find that cascading gratings with random ripple causes much less degradation than cascading gratings with the same ripple profile.     

Goal-directed imitation for robots: A bio-inspired approach to action understanding and skill learning

In this paper we present a robot control architecture for learning by imitation which takes inspiration from recent discoveries in action observation/execution experiments with humans and other primates. The architecture implements two basic processing principles: (1) imitation is primarily directed toward reproducing the outcome of an observed action sequence rather than reproducing the exact action means, and (2) the required capacity to understand the motor intention of another agent is based on motor simulation. The control architecture is validated in a robot system imitating in a goal-directed manner a grasping and placing sequence displayed by a human model. During imitation, skill transfer occurs by learning and representing appropriate goal-directed sequences of motor primitives. The robustness of the goal-directed organization of the controller is tested in the presence of incomplete visual information and changes in environmental constraints.     

Further improvement in high crystalline quality of homoepitaxial CVD diamond

High-quality homoepitaxial diamond (001) films with macroscopically flat surfaces have been successfully grown using a high-power microwave-plasma chemical-vapor-deposition (MWPCVD) method. In this study, further optimization of the homoepitaxial growth condition has been accomplished mainly by controlling off-angles to 5° along the <110> or <100> direction of high-pressure/high-temperature-synthesized Ib diamond (001) substrates. We have found that the homoepitaxial films deposited at reasonably high growth rates under the optimized growth condition including the off-angle of 3°–4° along the <110> direction have macroscopically flat surfaces, accompany very low or almost negligible densities of growth hillocks and yield strong free-exciton emissions in both steady-state cathodoluminescence and time-resolved photoluminescence spectra measured at room temperature. These indicate that apparent lateral growths suitable for high-quality homoepitaxial layers in the case of the high-power MWPCVD method, which are similar to those previously reported in the case of MWPCVD processes with low methane concentrations, rather quickly occur from step edges on (001) terraces and that they can be achieved more preferentially on the vicinal substrates at high temperatures and high methane concentrations.     

Improved Repeatability of Pulsed Field Magnetometry Measurements Through Peltier Effect Temperature Control of Measurement Coils

Pulsed Field Magnetometers (PFMs) offer a method for high speed full loop characterisation of hard magnetic materials. As there is no dependency on iron to close a circuit, the repeatability of the technique is good. By utilising Peltier effect techniques to control the temperature of the measurement coil systems employed in a Pulsed Field Magnetometer, the level of repeatability of measurements can be significantly improved.     

An efficient face segmentation algorithm based on binary partition tree

This paper presents an efficient face segmentation algorithm based on binary partition tree. Skin-like regions are first obtained by integrating the results of pixel classification and watershed segmentation. Facial features are extracted by the techniques of valley detection and entropic thresholding, and are used to refine the skin-like regions. In order to segment the facial regions from the skin-like regions, a novel region merging algorithm is proposed by considering the impact of the common border ratio between adjacent regions, and the binary partition tree is used to represent the whole region merging process. Then the facial likeness of each node in the binary partition tree is evaluated using a set of fuzzy membership functions devised for a number of facial primitives of geometrical, elliptical and facial features. Finally, an efficient algorithm of node selecting in the binary partition tree is proposed for the final face segmentation, which can exactly segment the faces without any underlying assumption. The performance of the proposed face segmentation algorithm is demonstrated by experimental results carried out on a variety of images in different scenarios.     

Admission and GoS control in a multiservice WCDMA system

We consider in this paper a WCDMA system with two types of calls: real time (RT) calls that have dedicated resources, and data non-real time (NRT) calls that are treated using a time-shared channel (such as the HDR or the HSDPA). We consider reservation of some resources for the NRT traffic and assume that this traffic is further assigned the resources left over from the RT traffic. The grade of service (GoS) of RT traffic is also controlled in order to allow for handling more RT calls during congestion periods, at the cost of degraded transmission rates. We consider both the downlink (with and without macrodiversity) as well as the uplink and study the blocking probabilities of RT traffic as well as the expected sojourn time of NRT traffic. We further study the conditional expected sojourn time of a data connection given its size and the state of the system. Finally, we extend our framework to handle handover calls.