Showing papers by "Harbin Institute of Technology published in 2018"

FFDNet: Toward a Fast and Flexible Solution for CNN-Based Image Denoising

Abstract: Due to the fast inference and good performance, discriminative learning methods have been widely studied in image denoising. However, these methods mostly learn a specific model for each noise level, and require multiple models for denoising images with different noise levels. They also lack flexibility to deal with spatially variant noise, limiting their applications in practical denoising. To address these issues, we present a fast and flexible denoising convolutional neural network, namely FFDNet, with a tunable noise level map as the input. The proposed FFDNet works on downsampled sub-images, achieving a good trade-off between inference speed and denoising performance. In contrast to the existing discriminative denoisers, FFDNet enjoys several desirable properties, including: 1) the ability to handle a wide range of noise levels (i.e., [0, 75]) effectively with a single network; 2) the ability to remove spatially variant noise by specifying a non-uniform noise level map; and 3) faster speed than benchmark BM3D even on CPU without sacrificing denoising performance. Extensive experiments on synthetic and real noisy images are conducted to evaluate FFDNet in comparison with state-of-the-art denoisers. The results show that FFDNet is effective and efficient, making it highly attractive for practical denoising applications.

Atomically dispersed manganese catalysts for oxygen reduction in proton-exchange membrane fuel cells

Abstract: Platinum group metal (PGM)-free catalysts that are also iron free are highly desirable for the oxygen reduction reaction (ORR) in proton-exchange membrane fuel cells, as they avoid possible Fenton reactions. Here we report an efficient ORR catalyst that consists of atomically dispersed nitrogen-coordinated single Mn sites on partially graphitic carbon (Mn-N-C). Evidence for the embedding of the atomically dispersed MnN4 moieties within the carbon surface-exposed basal planes was established by X-ray absorption spectroscopy and their dispersion was confirmed by aberration-corrected electron microscopy with atomic resolution. The Mn-N-C catalyst exhibited a half-wave potential of 0.80 V versus the reversible hydrogen electrode, approaching that of Fe-N-C catalysts, along with significantly enhanced stability in acidic media. The encouraging performance of the Mn-N-C catalyst as a PGM-free cathode was demonstrated in fuel cell tests. First-principles calculations further support the MnN4 sites as the origin of the ORR activity via a 4e− pathway in acidic media. Platinum group metal- and iron-free catalysts are highly desirable for the oxygen reduction reaction in proton-exchange membrane fuel cells. Now, Wu and co-workers show a carbon catalyst with atomically dispersed single Mn sites as an efficient catalyst with enhanced stability in acidic media.

Evaluating the replicability of social science experiments in Nature and Science between 2010 and 2015

Abstract: Being able to replicate scientific findings is crucial for scientific progress. We replicate 21 systematically selected experimental studies in the social sciences published in Nature and Science between 2010 and 2015. The replications follow analysis plans reviewed by the original authors and pre-registered prior to the replications. The replications are high powered, with sample sizes on average about five times higher than in the original studies. We find a significant effect in the same direction as the original study for 13 (62%) studies, and the effect size of the replications is on average about 50% of the original effect size. Replicability varies between 12 (57%) and 14 (67%) studies for complementary replicability indicators. Consistent with these results, the estimated true-positive rate is 67% in a Bayesian analysis. The relative effect size of true positives is estimated to be 71%, suggesting that both false positives and inflated effect sizes of true positives contribute to imperfect reproducibility. Furthermore, we find that peer beliefs of replicability are strongly related to replicability, suggesting that the research community could predict which results would replicate and that failures to replicate were not the result of chance alone.

Catalytic removal of aqueous contaminants on N-doped graphitic biochars: inherent roles of adsorption and nonradical mechanisms.

Abstract: Environmentally friendly and low-cost catalysts are important for the rapid mineralization of organic contaminants in powerful advanced oxidation processes (AOPs). In this study, we reported N-doped graphitic biochars (N-BCs) as low-cost and efficient catalysts for peroxydisulfate (PDS) activation and the degradation of diverse organic pollutants in water treatment, including Orange G, phenol, sulfamethoxazole, and bisphenol A. The biochars at high annealing temperatures (>700 °C) presented highly graphitic nanosheets, large specific surface areas (SSAs), and rich doped nitrogen. In particular, N-BC derived at 900 °C (N-BC900) exhibited the highest degradation rate, which was 39-fold and 6.5-fold of that on N-BC400 and pristine biochar, respectively, and the N-BC900 surpassed most popular metal or nanocarbon catalysts. Different from the radical-based oxidation in N-BC400/PDS via the persistent free radicals (PFRs), singlet oxygen and nonradical pathways (surface-confined activated persulfate–carbon compl...

Learning a Single Convolutional Super-Resolution Network for Multiple Degradations

Abstract: Recent years have witnessed the unprecedented success of deep convolutional neural networks (CNNs) in single image super-resolution (SISR). However, existing CNN-based SISR methods mostly assume that a low-resolution (LR) image is bicubicly downsampled from a high-resolution (HR) image, thus inevitably giving rise to poor performance when the true degradation does not follow this assumption. Moreover, they lack scalability in learning a single model to nonblindly deal with multiple degradations. To address these issues, we propose a general framework with dimensionality stretching strategy that enables a single convolutional super-resolution network to take two key factors of the SISR degradation process, i.e., blur kernel and noise level, as input. Consequently, the super-resolver can handle multiple and even spatially variant degradations, which significantly improves the practicability. Extensive experimental results on synthetic and real LR images show that the proposed convolutional super-resolution network not only can produce favorable results on multiple degradations but also is computationally efficient, providing a highly effective and scalable solution to practical SISR applications.

Dopant-induced electron localization drives CO 2 reduction to C 2 hydrocarbons

Abstract: The electrochemical reduction of CO2 to multi-carbon products has attracted much attention because it provides an avenue to the synthesis of value-added carbon-based fuels and feedstocks using renewable electricity Unfortunately, the efficiency of CO2 conversion to C2 products remains below that necessary for its implementation at scale Modifying the local electronic structure of copper with positive valence sites has been predicted to boost conversion to C2 products Here, we use boron to tune the ratio of Cuδ+ to Cu0 active sites and improve both stability and C2-product generation Simulations show that the ability to tune the average oxidation state of copper enables control over CO adsorption and dimerization, and makes it possible to implement a preference for the electrosynthesis of C2 products We report experimentally a C2 Faradaic efficiency of 79 ± 2% on boron-doped copper catalysts and further show that boron doping leads to catalysts that are stable for in excess of ~40 hours while electrochemically reducing CO2 to multi-carbon hydrocarbons

Cyclin D–CDK4 kinase destabilizes PD-L1 via cullin 3–SPOP to control cancer immune surveillance

Abstract: Treatments that target immune checkpoints, such as the one mediated by programmed cell death protein 1 (PD-1) and its ligand PD-L1, have been approved for treating human cancers with durable clinical benefit. However, many patients with cancer fail to respond to compounds that target the PD-1 and PD-L1 interaction, and the underlying mechanism(s) is not well understood. Recent studies revealed that response to PD-1-PD-L1 blockade might correlate with PD-L1 expression levels in tumour cells. Hence, it is important to understand the mechanistic pathways that control PD-L1 protein expression and stability, which can offer a molecular basis to improve the clinical response rate and efficacy of PD-1-PD-L1 blockade in patients with cancer. Here we show that PD-L1 protein abundance is regulated by cyclin D-CDK4 and the cullin 3-SPOP E3 ligase via proteasome-mediated degradation. Inhibition of CDK4 and CDK6 (hereafter CDK4/6) in vivo increases PD-L1 protein levels by impeding cyclin D-CDK4-mediated phosphorylation of speckle-type POZ protein (SPOP) and thereby promoting SPOP degradation by the anaphase-promoting complex activator FZR1. Loss-of-function mutations in SPOP compromise ubiquitination-mediated PD-L1 degradation, leading to increased PD-L1 levels and reduced numbers of tumour-infiltrating lymphocytes in mouse tumours and in primary human prostate cancer specimens. Notably, combining CDK4/6 inhibitor treatment with anti-PD-1 immunotherapy enhances tumour regression and markedly improves overall survival rates in mouse tumour models. Our study uncovers a novel molecular mechanism for regulating PD-L1 protein stability by a cell cycle kinase and reveals the potential for using combination treatment with CDK4/6 inhibitors and PD-1-PD-L1 immune checkpoint blockade to enhance therapeutic efficacy for human cancers.

Learning Spatial-Temporal Regularized Correlation Filters for Visual Tracking

Abstract: Discriminative Correlation Filters (DCF) are efficient in visual tracking but suffer from unwanted boundary effects. Spatially Regularized DCF (SRDCF) has been suggested to resolve this issue by enforcing spatial penalty on DCF coefficients, which, inevitably, improves the tracking performance at the price of increasing complexity. To tackle online updating, SRDCF formulates its model on multiple training images, further adding difficulties in improving efficiency. In this work, by introducing temporal regularization to SRDCF with single sample, we present our spatial-temporal regularized correlation filters (STRCF). The STRCF formulation can not only serve as a reasonable approximation to SRDCF with multiple training samples, but also provide a more robust appearance model than SRDCF in the case of large appearance variations. Besides, it can be efficiently solved via the alternating direction method of multipliers (ADMM). By incorporating both temporal and spatial regularization, our STRCF can handle boundary effects without much loss in efficiency and achieve superior performance over SRDCF in terms of accuracy and speed. Compared with SRDCF, STRCF with hand-crafted features provides a 5A— speedup and achieves a gain of 5.4% and 3.6% AUC score on OTB-2015 and Temple-Color, respectively. Moreover, STRCF with deep features also performs favorably against state-of-the-art trackers and achieves an AUC score of 68.3% on OTB-2015.

An Amorphous Noble-Metal-Free Electrocatalyst that Enables Nitrogen Fixation under Ambient Conditions.

Abstract: N2 fixation by the electrocatalytic nitrogen reduction reaction (NRR) under ambient conditions is regarded as a potential approach to achieve NH3 production, which still heavily relies on the Haber-Bosch process at the cost of huge energy and massive production of CO2 . A noble-metal-free Bi4 V2 O11 /CeO2 hybrid with an amorphous phase (BVC-A) is used as the cathode for electrocatalytic NRR. The amorphous Bi4 V2 O11 contains significant defects, which play a role as active sites. The CeO2 not only serves as a trigger to induce the amorphous structure, but also establishes band alignment with Bi4 V2 O11 for rapid interfacial charge transfer. Remarkably, BVC-A shows outstanding electrocatalytic NRR performance with high average yield (NH3 : 23.21 μg h-1 mg-1cat. , Faradaic efficiency: 10.16 %) under ambient conditions, which is superior to the Bi4 V2 O11 /CeO2 hybrid with crystalline phase (BVC-C) counterpart.

VITAL: VIsual Tracking via Adversarial Learning

Abstract: The tracking-by-detection framework consists of two stages, i.e., drawing samples around the target object in the first stage and classifying each sample as the target object or as background in the second stage. The performance of existing trackers using deep classification networks is limited by two aspects. First, the positive samples in each frame are highly spatially overlapped, and they fail to capture rich appearance variations. Second, there exists extreme class imbalance between positive and negative samples. This paper presents the VITAL algorithm to address these two problems via adversarial learning. To augment positive samples, we use a generative network to randomly generate masks, which are applied to adaptively dropout input features to capture a variety of appearance changes. With the use of adversarial learning, our network identifies the mask that maintains the most robust features of the target objects over a long temporal span. In addition, to handle the issue of class imbalance, we propose a high-order cost sensitive loss to decrease the effect of easy negative samples to facilitate training the classification network. Extensive experiments on benchmark datasets demonstrate that the proposed tracker performs favorably against state-of-the-art approaches.

Defect Engineering Metal‐Free Polymeric Carbon Nitride Electrocatalyst for Effective Nitrogen Fixation under Ambient Conditions

Abstract: Electrocatalytic nitrogen reduction reaction (NRR) under ambient conditions provides an intriguing picture for the conversion of N2 into NH3 . However, electrocatalytic NRR mainly relies on metal-based catalysts, and it remains a grand challenge in enabling effective N2 activation on metal-free catalysts. Here we report a defect engineering strategy to realize effective NRR performance (NH3 yield: 8.09 μg h-1 mg-1cat. , Faradaic efficiency: 11.59 %) on metal-free polymeric carbon nitride (PCN) catalyst. Illustrated by density functional theory calculations, dinitrogen molecule can be chemisorbed on as-engineered nitrogen vacancies of PCN through constructing a dinuclear end-on bound structure for spatial electron transfer. Furthermore, the N-N bond length of adsorbed N2 increases dramatically, which corresponds to "strong activation" system to reduce N2 into NH3 . This work also highlights the significance of defect engineering for improving electrocatalysts with weak N2 adsorption and activation ability.

Generative Adversarial Networks for Hyperspectral Image Classification

Abstract: A generative adversarial network (GAN) usually contains a generative network and a discriminative network in competition with each other. The GAN has shown its capability in a variety of applications. In this paper, the usefulness and effectiveness of GAN for classification of hyperspectral images (HSIs) are explored for the first time. In the proposed GAN, a convolutional neural network (CNN) is designed to discriminate the inputs and another CNN is used to generate so-called fake inputs. The aforementioned CNNs are trained together: the generative CNN tries to generate fake inputs that are as real as possible, and the discriminative CNN tries to classify the real and fake inputs. This kind of adversarial training improves the generalization capability of the discriminative CNN, which is really important when the training samples are limited. Specifically, we propose two schemes: 1) a well-designed 1D-GAN as a spectral classifier and 2) a robust 3D-GAN as a spectral–spatial classifier. Furthermore, the generated adversarial samples are used with real training samples to fine-tune the discriminative CNN, which improves the final classification performance. The proposed classifiers are carried out on three widely used hyperspectral data sets: Salinas, Indiana Pines, and Kennedy Space Center. The obtained results reveal that the proposed models provide competitive results compared to the state-of-the-art methods. In addition, the proposed GANs open new opportunities in the remote sensing community for the challenging task of HSI classification and also reveal the huge potential of GAN-based methods for the analysis of such complex and inherently nonlinear data.

Few-layer graphdiyne doped with sp-hybridized nitrogen atoms at acetylenic sites for oxygen reduction electrocatalysis.

Abstract: The oxygen reduction reaction (ORR) is a fundamental reaction for energy storage and conversion. It has mainly relied on platinum-based electrocatalysts, but the chemical doping of carbon-based materials has proven to be a promising strategy for preparing metal-free alternatives. Nitrogen doping in particular provides a diverse range of nitrogen forms. Here, we introduce a new form of nitrogen doping moieties —sp-hybridized nitrogen (sp-N) atoms into chemically defined sites of ultrathin graphdiyne, through pericyclic replacement of the acetylene groups. The as-prepared sp-N-doped graphdiyne catalyst exhibits overall good ORR performance, in particular with regards to peak potential, half-wave potential and current density. Under alkaline conditions it was comparable to commercial Pt/C, and showed more rapid kinetics. And although its performances are a bit lower than those of Pt/C in acidic media they surpass those of other metal-free materials. Taken together, experimental data and density functional theory calculations suggest that the high catalytic activity originates from the sp-N dopant, which facilitates O2 adsorption and electron transfer on the surface of the catalyst. This incorporation of chemically defined sp-N atoms provides a new synthetic route to high-performance carbon-based and other metal-free catalysts.

Multi-level Wavelet-CNN for Image Restoration

Abstract: The tradeoff between receptive field size and efficiency is a crucial issue in low level vision. Plain convolutional networks (CNNs) generally enlarge the receptive field at the expense of computational cost. Recently, dilated filtering has been adopted to address this issue. But it suffers from gridding effect, and the resulting receptive field is only a sparse sampling of input image with checkerboard patterns. In this paper, we present a novel multi-level wavelet CNN (MWCNN) model for better tradeoff between receptive field size and computational efficiency. With the modified U-Net architecture, wavelet transform is introduced to reduce the size of feature maps in the contracting subnetwork. Furthermore, another convolutional layer is further used to decrease the channels of feature maps. In the expanding subnetwork, inverse wavelet transform is then deployed to reconstruct the high resolution feature maps. Our MWCNN can also be explained as the generalization of dilated filtering and subsampling, and can be applied to many image restoration tasks. The experimental results clearly show the effectiveness of MWCNN for image denoising, single image super-resolution, and JPEG image artifacts removal.

Identification and Regulation of Active Sites on Nanodiamonds: Establishing a Highly Efficient Catalytic System for Oxidation of Organic Contaminants

Abstract: Nanodiamonds exhibit great potential as green catalysts for remediation of organic contaminants. However, the specific active site and corresponding oxidative mechanism are unclear, which retard further developments of high-performance catalysts. Here, an annealing strategy is developed to accurately regulate the content of ketonic carbonyl groups on nanodiamonds; meanwhile other structural characteristics of nanodiamonds remain almost unchanged. The well-defined nanodiamonds with well-controlled ketonic carbonyl groups exhibit excellent catalytic activity in activation of peroxymonosulfate for oxidation of organic pollutants. Based on the semi-quantitative and quantitative correlations of ketonic carbonyl groups and the reaction rate constants, it is conclusively determined that ketonic carbonyl groups are the catalytically active sites. Different from conventional oxidative systems, reactive oxygen species in nanodiamonds@peroxymonosulfate system are revealed to be singlet oxygen with high selectivity, which can effectively oxidize and mineralize the target contaminants. Impressively, the singlet-oxygen-mediated oxidation system significantly outperforms the classical radicals-based oxidation system in remediation of actual wastewater. This work not only provides a valuable insight for the design of new nanocarbon catalysts with abundant active sites but also establishes a very promising catalytic oxidation system for the green remediation of actual contaminated water.

Oxygen vacancy-rich 2D/2D BiOCl-g-C3N4 ultrathin heterostructure nanosheets for enhanced visible-light-driven photocatalytic activity in environmental remediation

Abstract: Photocatalytic degradation has been unearthed as a promising strategy for environmental remediation, and the calling is endless for more efficient photocatalytic system. In this study, a novel oxygen vacancy-rich two-dimensional/two-dimensional (2D/2D) BiOCl-g-C3N4 ultrathin heterostructure nanosheet (CN-BC) is successfully prepared by a facile solvothermal method for degradation of non-dye organic contaminants. HRTEM observes the formation of heterojunction, while ESR and XPS unveil the distinct oxygen vacancy concentrations. Density functional calculations reveal that the introduction of oxygen vacancies (OVs) brings a new defect level, resulting in the increased photoabsorption. Under visible light irradiation, the OVs-rich optimum ratio of CN-BC (50CN-50BC) Exhibits 95% removal efficiency of 4-chlorophenol within 2 h, which is about 12.5, 5.3 and 3.4 times as that of pure BiOCl, g-C3N4 and OVs-poor heterostructure, respectively. The photocatalytic mechanism of OVs-rich 50CN-50BC is also revealed, suggesting that the synergistic effect between 2D/2D heterojunction and oxygen vacancies greatly promotes visible-light photoabsorption and photoinduced carrier separation efficiency with a prolonged lifetime, which is confirmed by multiple optical and electrochemical analyses, including DRS, steady-state photoluminescence spectra, electrochemical impedance spectroscopy, photocurrent response and time-resolved fluorescence spectra. This study could bring new opportunities for the rational design of highly efficient photocatalysts by combining 2D/2D heterojunctions with oxygen vacancies in environmental remediation.

Automatic Pixel-Level Crack Detection and Measurement Using Fully Convolutional Network

Abstract: The spatial characteristics of cracks are significant indicators to assess and evaluate the health of existing buildings and infrastructures However, the current manual crack description

Is Sulfate Radical Really Generated from Peroxydisulfate Activated by Iron(II) for Environmental Decontamination

Abstract: It is well documented that the traditional Fenton reagent (ie, the combination of Fe(II) and H2O2) produces hydroxyl radical (•OH) under acidic conditions, while at near-neutral pH the reactive intermediate converts to ferryl ion (Fe(IV)) that can oxidize sulfoxides to produce corresponding sulfones, markedly differing from their •OH-induced products However, it remains unclear whether Fe(IV) is generated in the Fe(II) activated peroxydisulfate (PDS) process, where sulfate radical (SO4•-) is long recognized as the dominant intermediate in literature Here we demonstrated that SO4•- oxidized methyl phenyl sulfoxide (PMSO, a model sulfoxide) to produce biphenyl compounds rather than methyl phenyl sulfone (PMSO2) Interestingly, the formation of PMSO2 was observed when PMSO was treated by the Fe(II)/PDS system over a wide pH range, and the yields of PMSO2 were quantified to be ∼100% at acidic pH 3-5 The identification of Fe(IV) in the Fe(II)/PDS system could also reasonably explain the literature results on alcohol scavenging effect and ESR spectra analysis Further, a Fe(IV)-based kinetic model was shown to accurately simulate the experimental data This work urges re-evaluation of the Fe(II)/PDS system for environmental decontamination, given that Fe(IV) would have different reactivity toward environmental contaminants compared with SO4•- and/or •OH

The Unmanned Aerial Vehicle Benchmark: Object Detection and Tracking

Abstract: With the advantage of high mobility, Unmanned Aerial Vehicles (UAVs) are used to fuel numerous important applications in computer vision, delivering more efficiency and convenience than surveillance cameras with fixed camera angle, scale and view. However, very limited UAV datasets are proposed, and they focus only on a specific task such as visual tracking or object detection in relatively constrained scenarios. Consequently, it is of great importance to develop an unconstrained UAV benchmark to boost related researches. In this paper, we construct a new UAV benchmark focusing on complex scenarios with new level challenges. Selected from 10 hours raw videos, about 80, 000 representative frames are fully annotated with bounding boxes as well as up to 14 kinds of attributes (e.g., weather condition, flying altitude, camera view, vehicle category, and occlusion) for three fundamental computer vision tasks: object detection, single object tracking, and multiple object tracking. Then, a detailed quantitative study is performed using most recent state-of-the-art algorithms for each task. Experimental results show that the current state-of-the-art methods perform relative worse on our dataset, due to the new challenges appeared in UAV based real scenes, e.g., high density, small object, and camera motion. To our knowledge, our work is the first time to explore such issues in unconstrained scenes comprehensively. The dataset and all the experimental results are available in https://sites.google.com/site/daviddo0323/.

Review of Drug Repositioning Approaches and Resources.

Abstract: Drug discovery is a time-consuming, high-investment, and high-risk process in traditional drug development. Drug repositioning has become a popular strategy in recent years. Different from traditional drug development strategies, the strategy is efficient, economical and riskless. There are usually three kinds of approaches: computational approaches, biological experimental approaches, and mixed approaches, all of which are widely used in drug repositioning. In this paper, we reviewed computational approaches and highlighted their characteristics to provide references for researchers to develop more powerful approaches. At the same time, the important findings obtained using these approaches are listed. Furthermore, we summarized 76 important resources about drug repositioning. Finally, challenges and opportunities in drug repositioning are discussed from multiple perspectives, including technology, commercial models, patents and investment.

Waterproof and Tailorable Elastic Rechargeable Yarn Zinc Ion Batteries by a Cross-Linked Polyacrylamide Electrolyte

Abstract: Emerging research toward next-generation flexible and wearable electronics has stimulated the efforts to build highly wearable, durable, and deformable energy devices with excellent electrochemical performances. Here, we develop a high-performance, waterproof, tailorable, and stretchable yarn zinc ion battery (ZIB) using double-helix yarn electrodes and a cross-linked polyacrylamide (PAM) electrolyte. Due to the high ionic conductivity of the PAM electrolyte and helix structured electrodes, the yarn ZIB delivers a high specific capacity and volumetric energy density (302.1 mAh g–1 and 53.8 mWh cm–3, respectively) as well as excellent cycling stability (98.5% capacity retention after 500 cycles). More importantly, the quasi-solid-state yarn ZIB also demonstrates superior knittability, good stretchability (up to 300% strain), and superior waterproof capability (high capacity retention of 96.5% after 12 h underwater operation). In addition, the long yarn ZIB can be tailored into short ones, and each part sti...

End-to-End Blind Image Quality Assessment Using Deep Neural Networks

Abstract: We propose a multi-task end-to-end optimized deep neural network (MEON) for blind image quality assessment (BIQA). MEON consists of two sub-networks—a distortion identification network and a quality prediction network—sharing the early layers. Unlike traditional methods used for training multi-task networks, our training process is performed in two steps. In the first step, we train a distortion type identification sub-network, for which large-scale training samples are readily available. In the second step, starting from the pre-trained early layers and the outputs of the first sub-network, we train a quality prediction sub-network using a variant of the stochastic gradient descent method. Different from most deep neural networks, we choose biologically inspired generalized divisive normalization (GDN) instead of rectified linear unit as the activation function. We empirically demonstrate that GDN is effective at reducing model parameters/layers while achieving similar quality prediction performance. With modest model complexity, the proposed MEON index achieves state-of-the-art performance on four publicly available benchmarks. Moreover, we demonstrate the strong competitiveness of MEON against state-of-the-art BIQA models using the group maximum differentiation competition methodology.

A Novel Adaptive Kalman Filter With Inaccurate Process and Measurement Noise Covariance Matrices

Abstract: In this paper, a novel variational Bayesian (VB)-based adaptive Kalman filter (VBAKF) for linear Gaussian state-space models with inaccurate process and measurement noise covariance matrices is proposed. By choosing inverse Wishart priors, the state together with the predicted error and measurement noise covariance matrices are inferred based on the VB approach. Simulation results for a target tracking example illustrate that the proposed VBAKF has better robustness to resist the uncertainties of process and measurement noise covariance matrices than existing state-of-the-art filters.

DeepGauge: multi-granularity testing criteria for deep learning systems

Abstract: Deep learning (DL) defines a new data-driven programming paradigm that constructs the internal system logic of a crafted neuron network through a set of training data. We have seen wide adoption of DL in many safety-critical scenarios. However, a plethora of studies have shown that the state-of-the-art DL systems suffer from various vulnerabilities which can lead to severe consequences when applied to real-world applications. Currently, the testing adequacy of a DL system is usually measured by the accuracy of test data. Considering the limitation of accessible high quality test data, good accuracy performance on test data can hardly provide confidence to the testing adequacy and generality of DL systems. Unlike traditional software systems that have clear and controllable logic and functionality, the lack of interpretability in a DL system makes system analysis and defect detection difficult, which could potentially hinder its real-world deployment. In this paper, we propose DeepGauge, a set of multi-granularity testing criteria for DL systems, which aims at rendering a multi-faceted portrayal of the testbed. The in-depth evaluation of our proposed testing criteria is demonstrated on two well-known datasets, five DL systems, and with four state-of-the-art adversarial attack techniques against DL. The potential usefulness of DeepGauge sheds light on the construction of more generic and robust DL systems.

Initiating a mild aqueous electrolyte Co3O4/Zn battery with 2.2 V-high voltage and 5000-cycle lifespan by a Co(III) rich-electrode

Abstract: The Zn/Co3O4 battery is one of the few aqueous electrolyte batteries with a potential >2 V voltage. Unfortunately, so far, all reported Zn/Co3O4 batteries are using an alkaline electrolyte, resulting in poor cycling stability and environmental problems. Here, we report a Co(III) rich-Co3O4 nanorod material with vastly improved electrochemical kinetics. Zn/Co(III) rich-Co3O4 batteries can work well in ZnSO4 with a CoSO4 additive aqueous solution as a mild electrolyte, delivering a high voltage of 2.2 V, a capacity of 205 mA h g−1 (Co3O4) and an extreme cycling stability of 92% capacity retention even after 5000 cycles. Further mechanistic study reveals a conversion reaction between in situ formed CoO and Co3O4, which has never been observed in an alkaline Zn/Co3O4 battery. Subsequently, a flexible solid-state battery is constructed and reveals high flexibility and a high energy density of 360.8 W h kg−1 at a current density of 0.5 A g−1. Our research initiates the first Zn/Co3O4 battery working in a mild electrolyte, resulting in excellent electrochemical performance. It also indicates that the electrochemical kinetics can be effectively enhanced by fine tuning the atomic structure of electrode materials, opening a new door to improve the performance of aqueous electrolyte batteries.

A swarm of slippery micropropellers penetrates the vitreous body of the eye

Abstract: The intravitreal delivery of therapeutic agents promises major benefits in the field of ocular medicine. Traditional delivery methods rely on the random, passive diffusion of molecules, which do not allow for the rapid delivery of a concentrated cargo to a defined region at the posterior pole of the eye. The use of particles promises targeted delivery but faces the challenge that most tissues including the vitreous have a tight macromolecular matrix that acts as a barrier and prevents its penetration. Here, we demonstrate novel intravitreal delivery microvehicles—slippery micropropellers—that can be actively propelled through the vitreous humor to reach the retina. The propulsion is achieved by helical magnetic micropropellers that have a liquid layer coating to minimize adhesion to the surrounding biopolymeric network. The submicrometer diameter of the propellers enables the penetration of the biopolymeric network and the propulsion through the porcine vitreous body of the eye over centimeter distances. Clinical optical coherence tomography is used to monitor the movement of the propellers and confirm their arrival on the retina near the optic disc. Overcoming the adhesion forces and actively navigating a swarm of micropropellers in the dense vitreous humor promise practical applications in ophthalmology.

Shift-Net: Image Inpainting via Deep Feature Rearrangement

Abstract: Deep convolutional networks (CNNs) have exhibited their potential in image inpainting for producing plausible results. However, in most existing methods, e.g., context encoder, the missing parts are predicted by propagating the surrounding convolutional features through a fully connected layer, which intends to produce semantically plausible but blurry result. In this paper, we introduce a special shift-connection layer to the U-Net architecture, namely Shift-Net, for filling in missing regions of any shape with sharp structures and fine-detailed textures. To this end, the encoder feature of the known region is shifted to serve as an estimation of the missing parts. A guidance loss is introduced on decoder feature to minimize the distance between the decoder feature after fully connected layer and the ground-truth encoder feature of the missing parts. With such constraint, the decoder feature in missing region can be used to guide the shift of encoder feature in known region. An end-to-end learning algorithm is further developed to train the Shift-Net. Experiments on the Paris StreetView and Places datasets demonstrate the efficiency and effectiveness of our Shift-Net in producing sharper, fine-detailed, and visually plausible results. The codes and pre-trained models are available at https://github.com/Zhaoyi-Yan/Shift-Net.