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.     

Collaborative Distribution Alignment for 2D image-based 3D shape retrieval

Retrieving 3D shapes with 2D images has become a popular research area nowadays, and a great deal of work has been devoted to reducing the discrepancy between 3D shapes and 2D images to improve retrieval performance. However, most approaches ignore the semantic information and decision boundaries of the two domains, and cannot achieve both domain alignment and category alignment in one module. In this paper, a novel Collaborative Distribution Alignment (CDA) model is developed to address the above existing challenges. Specifically, we first adopt a dual-stream CNN, following a similarity guided constraint module, to generate discriminative embeddings for input 2D images and 3D shapes (described as multiple views). Subsequently, we explicitly introduce a joint domain-class alignment module to dynamically learn a class-discriminative and domain-agnostic feature space, which can narrow the distance between 2D image and 3D shape instances of the same underlying category, while pushing apart the instances from different categories. Furthermore, we apply a decision boundary refinement module to avoid generating class-ambiguity embeddings by dynamically adjusting inconsistencies between two discriminators. Extensive experiments and evaluations on two challenging benchmarks, MI3DOR and MI3DOR-2, demonstrate the superiority of the proposed CDA method for 2D image-based 3D shape retrieval task.     

Multi-scale and multi-patch transformer for sandstorm image enhancement

Sandstorm is a meteorological phenomenon common in arid and semi-arid regions. A sandstorm can carry large volumes of sand unexpectedly, which leads to severe color deviations and significantly degraded visibility when an image is taken in such a scenario. However, existing image enhancement methods cannot enhance sandstorm images well due to the challenging degradations and the scarcity of sandstorm training data. In this paper, we propose a Transformer with rotary position embedding to perform sandstorm image enhancement via building multi-scale and multi-patch dependencies. Our key insights in this work are 1) a multi-scale Transformer can globally eliminate the color deviations of sandstorm images via aggregating global information, 2) a multi-patch Transformer can recover local details well via learning the spatial variant degradations, and 3) a U-shape Transformer with rotary position embedding as the core unit of multi-scale and multi-patch Transformer can effectively build the long-range dependencies. We also contribute a real-world Sandstorm Image Enhancement (SIE) dataset including 1,400 sandstorm images with different degrees of degradations and various scenes. Experiments performed on synthetic images and real-world sandstorm images demonstrate that our proposed method not only obtains visually pleasing results but also outperforms state-of-the-art methods qualitatively and quantitatively.     

Influencer-defaulter mutation-based optimization algorithms for predicting electricity prices

Efficient electricity price forecasting plays a significant role in our society. In this paper, a novel influencer-defaulter mutation (IDM) mutation operator has been proposed. The IDM operator has been combined with six well-known optimization algorithms to create mutated optimization algorithms whose performance has been tested on twenty-four standard benchmark functions. Further, the artificial neural network is integrated with mutated optimization algorithms to solve the electricity price prediction problem. The policymakers can identify appropriate variables based on the predicted prices to help future market planning. The statistical results prove the efficacy of the IDM operator on the recent optimization algorithms.     

Deep Learning based Early Prediction Scheme for Breast Cancer

Wireless Personal Communications - Breast cancer is one of the rapid spreading diseases resulting in the death of younger age group of women. Unfortunately, as the detection of cancer is at later...     

Stabilization for Schrödinger Equation with Internal Damping and Boundary Disturbance

In this paper, the stabilization for Schrödinger equation subject to internal damping and boundary disturbance at the control end is investigated. Due to its immeasurability, the nonlinear observer system is designed to obtain the state information, and the existence of weak solution and its convergence for the nonlinear observer system are proved. The feedback control is realized by the backstepping transformation. In addition, the adaptive disturbance rejection control is applied to estimate the disturbance. According to the observer and disturbance estimation, the feedback control is finally designed to stabilize the system asymptotically.

     

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.     

Importance of optical homogeneity for high-quality transparent ceramics

Two types of transparent Y₂O₃ ceramics without including large scattering sources such as residual pores, one with very high optical homogeneity (type A) and another one with slightly insufficient optical homogeneity (type B), are purposely prepared, and their optical properties are investigated and compared qualitatively and quantitatively. Type A ceramic exhibits transmittance characteristics with very low internal loss in the visible to infrared wavelength region, while type B ceramic is inferior in various optical performances especially in the short (visible) wavelength region. In type B ceramic, birefringence occurs due to optical inhomogeneity in the visible region, resulting in a decrease in the extinction ratio. Non-uniform refractive index distribution is also observed in the Schlieren image of type B ceramic, hence the laser beam quality through that material is degraded. This study proved the importance of optical homogeneity of transparent ceramics and clarified the problems in actual applications.     

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。