Search

Recent years deep learning-based methods incorporating facial prior knowledge for face super-resolution (FSR) are advancing and have gained impressive performance. However, some important priors such as facial landmarks are not fully exploited in existing methods, leading to noticeable artifacts in the resultant SR face images especially under large magnification. In this paper, we propose a novel multi-level landmark-guided deep network (MLGDN) for FSR. More specifically, to fully exploit the dependencies between low and high resolution images and to reduce network parameters as well as capture more reliable feature representation, we introduce a recursive back-projection network with a particular feedback mechanism for coarse-to-fine FSR. Furthermore, we incorporate an attention fusion module in the front of backbone network to strengthen face components and a feature modulation module to refine features in the middle of backbone network. By this way, the facial landmarks extracted from face images can be fully shared by the modules in different levels, which benefit to produce more faithful facial details. Both quantitative and qualitative performance evaluations on two benchmark databases demonstrate that the proposed MLGDN can achieve more impressive SR results than other state-of-the-art competitors. Code will be available at https://github.com/zhuangcheng31/MLG_Face.git/.

Carbon black (CB) has been demonstrated to have adverse effects on the lung tissue. Few studies explored the effects of CB on the cerebellum, widely recognized to contribute to gait and balance coordination and timing in the motor domain. Some studies have reported that inflammatory response and damaged autophagy are important mechanisms of CB toxicity and can be repaired after the recovery. The present study aimed to determine whether long-term CB exposure could induce the inflammation and damaged autophagy of the cerebellum. The rats were randomly divided into four groups. The control group received the filtered air for 90 days; the carbon black (CB) group received CB particles for 90 days; the recovery (R) group received CB for 90 days and recovered for another 14 days; the recovery control (RC) group received filtered air for 104 days. The purpose of the R group was to test whether neuroinflammation and autophagy could be repaired after short-term recovery. The western blot and immunohistochemistry revealed that long-term CB exposure induced augmented level of pro-inflammatory cytokines (Interleukin-1β, IL-1β; Interleukin-6, IL-6; and Tumor Necrosis Factor-α, TNF-α) and anti-inflammatory cytokine (Interleukin-10, IL-10). The autophagic markers (Beclin1 and LC3) were increased in both CB group and R group. These findings clearly demonstrated that long-term CB exposure induced inflammation and autophagy in the cerebellum, which were not obviously improved after short-term recovery.

Surface morphology measurement based on multiple wavelengths interferometric theory has been widely used, and the alignment error of initial phase of different wavelengths can influence the measurement accuracy. A phase shift method to drive the piezoelectric transducer (PZT) by zigzag with a fallback path is proposed, which can overcome the hysteresis and creep behavior of the PZT before switching to different wavelengths. The ellipse fitting and unwrapping algorithms are applied to extract phase-shifts from interferograms. The starting points of different wavelengths are aligned according to the displacement and movement direction of PZT. The experimental results show that the presented method can improve the aligned accuracy of the initial position during multiple-wavelength microscopic interferometry measurement.

Generalized stacking fault energy (GSFE) is an important property in understanding the plastic deformations of metals. However, the traditional way to calculate it one by one is not efficient, this work introduces a high-throughput workflow to calculate GSFEs using density functional theory (DFT) calculations and climbing image-nudged elastic band (CI-NEB) method. Based on open-source computational tools from the Materials Project infrastructure, this computation framework automates the procedure of building perfect and faulted slab models with certain orientations, performing DFT simulations and extracting the results into database. The computed GSFEs from this work and ductility parameter based on the GSFE are consistent with reported data from previous literatures, validating the accuracy of our results and algorithm. Such a GSFE workflow may speed up the development and understanding of the mechanical properties of metals or alloys by enabling the computations of GSFEs in an automatic and high-throughput way.

This study aimed to determine the effects of Enrofloxacin (ENR) exposure and depuration on the disruption of thyroid function and growth of juvenile grass carp (Ctenopharyngodon idella) as well as to assess the risk of ENR exposure to human health. Juvenile grass carp were treated with ENR solutions at different concentration gradients for 21 days and then depurated for 14 days. The results indicated ENR accumulation in the juvenile grass carp muscles, which persisted after depuration. In addition, exposure to ENR could alter growth by regulating the expression of genes associated with growth hormone/insulin-like growth factor (GH)/IGF) axis and the hypothalamic-pituitary-thyroid (HPT) axis. During ENR exposure, no significant changes in growth hormone levels were observed; however, a significant increase in the growth hormone level was noted. GH/IGF axis-related genes were upregulated after ENR exposure, and their expression levels remained high after depuration. Notably, a significant increase in the serum triiodothyronine (T3) and thyroxine (T4) levels coincided with the upregulation of HPT axis-related genes in both exposure and depuration treatments, and their expression levels remained high after depuration. Therefore, juvenile grass carp exposure to ENR induces physiological stress through HPT and GH/IGF axes that cannot be recovered after depuration. ENR accumulates in the muscles of juvenile grass carp and may pose a threat to human health. Therefore, exposure of juvenile grass carp to ENR results in impaired thyroid function and impaired growth. In addition, consumption of ENR-exposed fish poses human health risks.

In this study, Chinese rare minnow (Gobiocypris rarus) larvae were exposed to the control group, Cd concentrations (0.5 and 2.5 mg/L), and Hg concentrations (0.1 and 0.3 mg/L) for 96 h. Transcriptome analysis showed that 816 and 1599 significantly differentially expressed genes (DEGs) were identified in response to 2.5 mg/L Cd

The present study aimed to evaluate the effects of using different feeding strategies on the growth performance, feed efficiency, antioxidant capacity, digestive enzyme activities, and serum biochemical parameters of common carp, Cyprinus carpio for 40 days under high-temperature stress. A total of 120 fish (average initial weight 25.30 ± 1.27 g) were distributed in 12 tanks maintained in a circulating aquarium system (a water flow rate of 1 L/min). Fish were divided into four treatments: feeding every day with two meals, feeding one-day intervals with two meals, feeding two-day intervals with two meals, and feeding four-day intervals with two meals (named as T0, T1, T2, and T3, respectively). The results show that the weight gain (WG) and specific growth rate (SGR) of the T2 group were significantly higher than the other groups, while the feed conversion rate (FCR) of the T2 group was lower than in the other groups (P

The effect of the differential speed rolling on the ridging resistance and the crystallographic texture evolution of ultra purified 17%Cr ferritic stainless steel was investigated in order to select the optimal procedure for improving the surface quality. In the present study, the three different cold rolling processes were carried out: a conventional cold rolling and two different speed ratios asymmetric rolling. In the conventional cold rolling, the ridging of ferritic stainless steel was pronounced due to the band-like clusters of grains with similar crystallographic orientations. For differential speed rolling, ridging depth decreased and ridging was less pronounced when a higher different speed ratio was applied. The results show that severe ridging is attributed to the presence of band-like clusters of γ-fiber (〈111〉//ND) oriented grains. The shear bands introduced by differential speed rolling can enhance the nucleation for recrystallization both in grain interiors and grain boundaries, which can fragment the band-like clusters of similar orientation effectively, promoting the random grain orientation and improving surface quality.

In order to reduce cost and improve service reliability, self-organizing networks (SON) features are being introduced gradually with the arrival of new 4G systems in radio access networks. Cell outage compensation (COC) is one of the most important tasks in the context of SON. This paper demonstrates a self-organized approach for COC, which based on fuzzy Q-learning and operates in a fully autonomous manner. To improve the effectiveness of the compensation algorithm and reduce the complexity, a method of selecting the compensation cells based on knowledge of cell load and radiated power level respectively is proposed. Then the joint downtilt and transmit power compensation is described as a multi-agent reinforcement learning problem and solved using fuzzy Q-Learning (FQL). The experiment results show that the network performance degradation is minimized when a cell is in outage through compensation. Moreover, the hybrid approach which adjusts downtilt and power simultaneously delivers higher performance than the strategies varied in antenna tilt or transmit power only.

Introduction/Background The objective of this study is to test the clinical effectiveness of a synergy-based multi-muscle functional electrical stimulation (FES) strategy to accelerate post-stroke motor function recovery via a new task-oriented training protocol. Material and method A randomized, single-blind, placebo controlled protocol was designed and tested in this study. Daily FES interventions were applied to patients, approximately one hour each day for 5 consecutive days. Stimulation parameters were determined for each patient using an algorithm based on muscle-synergies obtained from healthy subjects. The task-oriented training protocol required patients to repeat forward and lateral reaching movements, with concurrent assistance from FES. Sham interventions were tested with all FES electrodes attached but without actual stimulations.20 patients with ischemic post-stroke hemiparesis participated in this study. FES interventions were applied to 10 patients, and 10 patients received sham interventions. We measured kinematics, electromyography, and upper-extremity Fugl-Meyer scores of patients before and after intervention. Results After 5 days of FES intervention, all patients showed increased Fugl-Meyer scores and peak movement velocity. Interestingly, the post-intervention muscle synergy in patients became more similar to that of healthy controls. No comparable improvements were observed after sham interventions. Particularly, a self-comparison in the patient received both sham and FES interventions highlighted the effectiveness of FES with task-oriented training. Conclusion Our results provided early evidence that there is clear clinical benefit to concurrently stimulate multiple muscles to assist residual movements using FES, and synergy-based FES induces restoration of more normal kinematics and muscle activation pattern. The FES and task-oriented training protocol showed initial indications of effectiveness with a short-term intervention.