Mutual information maximizing GAN inversion for real face with identity preservation

Recent generative adversarial networks (GANs) have yielded remarkable performance in face image synthesis. GAN inversion embeds an image into the latent space of a pretrained generator, enabling it to be used for real face manipulation. However, current inversion approaches for real faces suffer the dilemma of initialization collapse and identity loss. In this paper, we propose a hierarchical GAN inversion for real faces with identity preservation based on mutual information maximization. We first use a facial domain guaranteed initialization to avoid the initialization collapse. Furthermore, we prove that maximizing the mutual information between inverted faces and their identities is equivalent to minimizing the distance between identity features from inverted and original faces. Optimization for real face inversion with identity preservation is implemented on this mutual information-maximizing constraint. Extensive experimental results show that our approach outperforms state-of-the-art solutions for inverting and editing real faces, particularly in terms of face identity preservation.     

A review of thin film electrolytes fabricated by physical vapor deposition for solid oxide fuel cells

The ohmic resistance in solid oxide fuel cells (SOFCs) mainly comes from the electrolyte, which can be reduced by developing novel electrolyte materials with higher ionic conductivity and/or fabricating thin-film electrolytes. Among various kinds of thin-film fabrication technology, the physical vapor deposition (PVD) method can reduce the electrolyte thickness to a few micrometers and mitigate the issues associated with high-temperature sintering, which is necessary for wet ceramic methods. This review summarizes recent development progress in thin-film electrolytes fabricated by the PVD method, especially pulsed laser deposition (PLD) and magnetron sputtering. At first, the importance of the substrate surface morphology for the quality of the film is emphasized. After that, the fabrication of thin-film doped-zirconia and doped-ceria electrolytes is presented, then we provide a brief summary of the works on other types of electrolytes prepared by PVD. Finally, we have come to the summary and made perspectives.     

Phase composition and property evolution of (Yb1−xHox)2Si2O7 solid solution as environmental/thermal barrier coating candidates

RE disilicates are good candidates as environmental/thermal barrier coating for SiC_f/SiC composite in harsh gas turbine engines. We designed (Yb_1?xHo_x)₂Si₂O₇ solid solutions and studied mechanical properties, thermal properties, and water vapor resistance. Powders with different compositions were synthesized by pressureless sintering, and bulk samples were prepared by Spark Plasma Sintering (SPS). Polymorphic changes with temperature and composition of the solid solutions were examined. Through doping Ho into Yb₂Si₂O₇, water vapor corrosion resistance is significantly promoted, and thermal expansion coefficient is maintained close to that of Si-based ceramics. Compared with host disilicates, thermal conductivity of solid solutions are decreased, and mechanical properties, including Vickers hardness and fracture toughness, are increased. A two-phase domain is found at (Yb_1/2Ho_1/2)₂Si₂O₇, and the γ to δ phase transition of Ho₂Si₂O₇ is observed during SPS. Among all samples, γ-(Yb_1/3Ho_2/3)₂Si₂O₇ possesses superior high temperature stability, and excellent water vapor resistance, indicating its performance as environmental/thermal barrier coating.     

Rip currents near coastal structures in Lake Michigan: Characterization and assessment for warnings

Rip currents near coastal structures commonly occur in Lake Michigan in the Great Lakes region of the United States. Lack of timely warning due to undocumented characteristics of rip currents and no assessment tool can contribute to tragic drownings incidents. In this paper, we characterized rip current occurrences near breakwater structures and developed an assessment tool for providing timely rip current warnings to beachgoers at the study site, City of Port Washington, WI. Characteristics of rip currents near the structure were observed from field measurements or visual images. Deflection rip currents had speeds of ～ 0.2 m/s and lasted for several hours. The rip current occurrences were associated with environmental proxies. It was found that rip currents can occur even when the water appears calm near the structure. A Structure Rip Checklist and Assessment Matrix (SRiCAM) with a four-tiered risk was developed and validated using observations. Furthermore, the SRiCAM was integrated into cyberinfrastructure with a data contingency plan to provide real-time warnings to the public. The applicability of the SRiCAM to other locations across Lake Michigan was further tested and results are promising. Overall, the SRiCAM has the potential to be widely extended to foster recreational water safety and resilience to rip current hazards in the Great Lakes.     

The effect of rare-earth oxides on the energy storage performances in BaTiO3 based ceramics

In this work, 0.7BaTiO₃-0.3Sr_0.2Bi_0.7TiO₃ (0.7BT-0.3SBT) ceramics with 0.15 mol% various rare-earth oxides doped are designed and synthesized by the conventional solid-state route. All prepared samples exhibited a single perovskite phase and dense microstructure with fine grain size (0.2–0.5 μm) after sintering at 1180 °C. Especially, the Gd-doped 0.7BT-0.3SBT ceramics exhibited excellent energy storage performances; the corresponding recoverable energy density and efficiency were 3.2 J/cm³ and 91.5% under an electric field of 330 kV/cm, respectively. Meanwhile, doping with Gd caused the BT-based ceramics to possess excellent temperature (30–150 °C) and outstanding frequency stabilities (10–1000 Hz). Moreover, the pulsed charge-discharge experiments revealed that a high power density of 59 MW/cm³ and a fast discharge speed of 110 ns with outstanding temperature stability could be synchronously obtained in the Gd-doped composition. All these features are attractive for pulsed power applications.     

Mg gas infiltration for the fabrication of MgB2 pellets using nanosized and microsized B powders

MgB₂ superconductor pellets were synthesized through Mg gas infiltration method using nanosized- and microsized B powders. There was a marked difference in the superconducting properties of the two samples, particularly in the pinning force and dominant pinning mechanism. The microstructures of the samples were observed using HR-TEM and STEM-HAADF, and the results showed that the primary reason for the difference in the superconducting properties is the distribution of the nanosized second-phase particle MgO. Additionally, a feasible reaction model for the Mg gas infiltration method was established. Compared to the Mg liquid infiltration method, the gas infiltration showed better penetrability ability with a small amount of residual Mg. This study presents a novel synthesis process to fabricate an MgB₂ pellet with superior density and superconducting properties. This method can be used in multiple applications such as superconducting bearings, compact superconductor magnets, and magnetic shielding.     

A kinetic model involving oxidation and volatilization in SiC-B4C-xAl2O3 ceramics system

In this paper, a new kinetic model considering both oxidation and volatilization kinetics is established and applied to analyze the oxidation of SiC-B₄C-xAl₂O₃ ceramics and other systems in various oxidation conditions. The effects of diffusion area and volume changes during the oxidation process are considered in this model. The physical meaning of each parameter in this model is explicit and simple. According to this model, the diffusion coefficient of species and the corresponding diffusion activation energy are easily available. The practicability of this model is well verified by the experimental data of SiC-B₄C-xAl₂O₃ and other systems oxidized under different conditions. In addition, the practice shows that the model is applicable not only to the systems where oxidation and volatilization coexist, but also to the system where only oxidation plays a major role. We hope the model proposed in this work can be used in other materials with more complex environments.     

基于卷积神经网络的5G蜂窝网络无线定位方法

5G蜂窝网络发展迅猛，其覆盖面积将逐渐增大，因此使用5G蜂窝网络进行定位是有研究潜力的研究方向。本文提出一种新的深度学习技术来实现高效、高精度和低占用的定位，以代替传统指纹定位过程中繁重的指纹库生成以及距离计算。该方法建立了一个特殊的卷积神经网络，并根据5G天线信号的接收信号强度指示、相位和到达角等特征量，选择合适的输入数据格式构造样本组建训练集，对该卷积神经网络进行训练。训练得到的卷积神经网络可以替代指纹定位中的庞大指纹库，非常有利于直接在5G移动设备端实现定位。虽然卷积神经网络在训练过程中需要大量时间，但在训练完毕后直接进行分类定位的速度非常快，可以保障定位实现的实时性。本文所实现的卷积神经网络权重与偏置所占内存不到0.5 MB，且能够在实际应用环境中以95%的定位准确率以及0.1 m的平均定位精度实现高精度定位。     

DFD-SS: Document Forgery Detection using Spectral – Spatial Features for Hyperspectral Images

In the present era of machines and edge-cutting technologies, still document frauds persist. They are done intuitively by using almost identical inks, that it becomes challenging to detect them—this demands an approach that efficiently investigates the document and leaves it intact. Hyperspectral imaging is one such a type of approach that captures the images from hundreds to thousands of spectral bands and analyzes the images through their spectral and spatial features, which is not possible by conventional imaging. Deep learning is an edge-cutting technology known for solving critical problems in various domains. Utilizing supervised learning imposes constraints on its usage in real scenarios, as the inks used in forgery are not known prior. Therefore, it is beneficial to use unsupervised learning. An unsupervised feature extraction through a Convolutional Autoencoder (CAE) followed by Logistic Regression (LR) for classification is proposed (CAE-LR). Feature extraction is evolved around spectral bands, spatial patches, and spectral-spatial patches. We inspected the impact of spectral, spatial, and spectral-spatial features by mixing inks in equal and unequal proportion using CAE-LR on the UWA writing ink hyperspectral images dataset for blue and black inks. Hyperspectral images are captured at multiple correlated spectral bands, resulting in information redundancy handled by restoring certain principal components. The proposed approach is compared with eight state-of-art approaches used by the researchers. The results depicted that by using the combination of spectral and spatial patches, the classification accuracy enhanced by 4.85% for black inks and 0.13% for blue inks compared to state-of-art results. In the present scenario, the primary area concern is to identify and detect the almost similar inks used in document forgery, are efficiently managed by the proposed approach.     

(Cu1/3Nb2/3)4+ ionic substituting effects,low-temperature sintering behaviors,and improved temperature stability of Ba3Nb4Ti4O21 microwave dielectric ceramics

In this study, (Cu_1/3Nb_2/3)⁴⁺ complex cation and BaO–ZnO–B₂O₃ glass frit were adopted to solve the high sintering temperature and poor temperature stability of Ba₃Nb₄Ti₄O₂₁ ceramics. It is shown that pure Ba₃Nb₄Ti₄O₂₁ phase was formed when Ti site was partially replaced by (Cu_1/3Nb_2/3)⁴⁺ cation. The increasing number of dopants decreases the dielectric polarizability, correspondingly, the dielectric constant and temperature coefficient of the resonance frequency values are reduced consistently. The variation of the Q × f value is determined by internal ionic packing fraction and external sintering densification. The (Cu_1/3Nb_2/3)⁴⁺ cation effectively decreases the suitable sintering temperature from 1200 to 1050 °C while greatly improving the temperature stability. BaO–ZnO–B₂O₃ glass was used to further improve the low-temperature sintering characteristics of Ba₃Nb₄Ti₄O₂₁ ceramics. It is proven that the addition of glass frits effectively decreases the temperature to 925 °C with combinational excellent microwave dielectric properties: ε_r ～55.6, Q × f ～5700 GHz, τ_f ～3 ppm/^°C, making the Ba₃Nb₄Ti₄O₂₁ ceramics promising in the applications of low-temperature cofired ceramic technology.