Search

To explain the factors behind the regime resilience in China, this article focuses on the foundational and institutional resources that the Chinese Communist Party (CCP) accumulated in the earlier stages of regime development and their lasting influence on regime trajectory. At the stage of regime foundation, the CCP successfully constructed the founding myth of the party-state as the foundational resource to legitimize its rule. At the later stage of regime adaptation, institutional adaptation during the process of achieving modernization allowed the CCP to accumulate substantial institutional resources to further buttress the regime. While negotiating the space between state and society, the CCP regime demonstrated high capacity in granting more autonomy to market actors to accelerate growth while increasing the embeddedness of economic strata in the system through informal and formal institutional arrangements. When confronted with a serious crisis, the CCP was able to draw strength from the foundational and institutional resources to survive. This article argues that the CCP's efforts in accumulating strength at the early stages of regime development continue to exert strong influence on China's regime trajectory at later stages. [ABSTRACT FROM AUTHOR]

The tracking performance of Multi-Object Tracking (MOT) has recently been improved by using discriminative appearance and motion features. However, dense crowds and occlusions significantly reduce the reliability of these features, resulting in unsatisfied tracking performance. Thus, we design an end-to-end MOT model based on Graph Convolutional Neural Networks (GCNNs) which fuses four classes of features that characterize objects from their appearances, motions, appearance interactions, and motion interactions. Specifically, a Re-Identification (Re-ID) module is used to extract more discriminative appearance features. The appearance features from object tracklets are then averaged to simplify the proposed tracker. Then, we design two GCNNs to better distinguish objects. One is for extracting interactive appearance features, and the other is for interactive motion features. A fusion module then fuses those features, getting the global feature similarity based on which an association component calculates the MOT matching results. Finally, we semantically visualize relevant structures with the GNNExplainer for insight into the proposed tracker. The evaluation results on MOT16 and MOT17 benchmarks show that our model outperforms the state-of-the-art online tracking methods in terms of Multi-Object Tracking Accuracy and Identification F1 score which is consistent with the results from the GNNExplainer. [ABSTRACT FROM AUTHOR]

This study constructs a composite indicator system covering the core dimensions of medical equipment input and output. Based on this system, an innovative cone-constrained data envelopment analysis (DEA) model is designed. The model integrates the advantages of the analytic hierarchy process (AHP) with an improved criterion importance through intercriteria correlation (CRITIC) method to determine subjective and objective weights and employs game theory to obtain the final combined weights, which are further incorporated as constraints to form the cone-constrained DEA model. Finally, a bidirectional long short-term memory (Bi-LSTM) model with an attention mechanism is introduced for integration, aiming to provide a novel and practical model for evaluating the effectiveness of medical equipment. The proposed model has essential reference value for optimizing medical equipment management decision-making and investment strategies. [ABSTRACT FROM AUTHOR]

In 2018, 43 leading International Relations scholars in the United States signed a public statement in support of an urgent call to preserve the current international order, triggering heated scholarly debates. The idealized form of the liberal international order was criticized by many scholars for its chronic problems, including the contradictions between proclaimed liberal values and illiberal behaviors, the inability to reform its institutional pillars to accommodate the diverse group of emerging powers, and the tensions between the defenders of this order and its challengers. These problems became fully exposed under the external shock caused by the coronavirus pandemic. As the coronavirus spreads globally and disrupts the world's political, economic, and social fabric, several forces that have gained momentum and strength during the last decade are now converging as a formidable force that may reconfigure the post-pandemic international order. This paper addresses three significant challenges to the foundations of the current liberal order: (1) the entrenchment of authoritarianism, characterized by authoritarian resilience, autocratization, and the consolidation of competing authoritarian political-economic models; (2) the exacerbation of nationalism enabled by nationalist and populist politicians; and (3) the intensified competition among major powers. China has played mixed roles in the process of reconfiguring the current order. It challenges the mythologized liberal international order and exposes the contradictions in the dominant Western model, while promoting an alternative hybrid political-economic model. The shock brought by the pandemic has provided ample opportunities for China to extend its networks and expand international space for its model. [ABSTRACT FROM AUTHOR]

The phosphorus (P) modified MoO-BiSiO/SiO catalyst was prepared by a simple co-impregnation method and investigated in the epoxidation of propylene by molecular oxygen. The catalyst was characterized by X-ray diffraction (XRD), N adsorption-desorption analysis, NH-temperature-programmed desorption (NH-TPD), transmission electron microscopy, and Raman spectroscopy. It was found that the P-modified MoO-BiSiO/SiO catalyst with a P/Mo molar ratio of 0.5 exhibits the best catalytic performance for epoxidation of propylene by O, the TOFs for propylene oxide (PO) formation was four times higher than that of the unmodified one at 633 K. The modification by P could promote the dispersion of MoO nanoparticles and increase the number of weak and moderate acid sites with respect to the phosphorus-free MoO-BiSiO/SiO catalyst, which were beneficial to the formation of PO. Moreover, the introduction of P also could protect the mesoporous structure by inhibiting the formation of BiMoO, which was beneficial to the dispersion of active species. We suppose that the phosphorus, bismuth and molybdenum species of P-modified MoO-BiSiO/SiO catalyst play important roles for propylene epoxidation by molecular oxygen. [ABSTRACT FROM AUTHOR]

The use of local features for image representation has become popular in recent years. Local features are often used in the bag-of-visual-words scheme. Although proven effective, this method still has two drawbacks. First, local regions from which local features are extracted are not discriminative enough for visual tasks. Hence, the combination of local features is necessary. Second, the semantic gap between visual features and human perception also hinders the performance. To address these two problems, in this paper, we propose a novel contextual exemplar classifier-based method for image representation and apply it for classification tasks. Each exemplar classifier is trained to separate one training image from the other images of different classes. We partition each image into a number of regions and use the responses of these exemplar classifiers as the image region’s representation. The contextual relationship is then modeled using mixture Dirichlet distributions. A bilayer model is used to predict image classes with L2 constraints. Experimental results on the Natural Scene, Caltech-101/256, Flower-17/102, and SUN-397 data sets show that the proposed method is able to outperform the state-of-the-art local feature-based methods for image classification. [ABSTRACT FROM PUBLISHER]

In this paper, we report the divalent ions Mg, Zn, and Cu as dopants to manipulate the crystal field asymmetry in a new cubic YYbErAlF phase which features two different coordination sites for rare earth ions. X-ray powder diffraction, transmission electron microscopy, X-ray photoelectron spectroscopy, and fluorescence spectrophotometer were employed for the crystal structure analysis and luminescence performance investigation. Results reveal that the phase transition from cubic to orthogonal of YYbErAlF is promoted by Mg doping. Among Mg, Zn, and Cu, Mg is found to be the most effective dopant for the upconversion performance enhancement. The crystal lattice structure asymmetry increases with the rising of Mg doping concentration and reaches the peak when 6 mol% Mg is introduced; correspondingly, the brightness of 408 and 654 nm emissions are strikingly enhanced over 20 times, and the lifetime of 540 nm emission is shortened to half of that in the Mg-free sample. The investigation results establish the understanding of divalent ions as dopants for adjusting upconversion luminescence performance and may be helpful for researchers to develop quick responsive upconversion luminescence materials. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new online learning method for measuring affinity between tracklets in multi-target tracking. As targets and background usually keep changing in the video, fixed affinity measurement could not adapt to their variations. Most existing affinity learning methods construct labeled samples based on the obtained tracklets, and then minimize a predefined loss function to get an optimal affinity measurement. However, those methods simply assume that the training error equals to testing error which is not true in many of real time tracking scenarios. Differently, we propose to learn affinity measurement through CovBoosting, which considers the evolution of the tracklets and could obtain affinity measurement with more discriminative ability. To deal with targets׳ disappearance and new targets׳ appearance, we combine tracklet affinity with contextual information to do an optimal inference. Moreover, an online updating algorithm is developed to guarantee that the learned tracklet affinity is always optimal for tracking targets in current sliding window. Experimental results on benchmark datasets demonstrate that tracklet affinity learned with our method is more discriminative and could greatly improve the performance of the multi-target tracker. [ABSTRACT FROM AUTHOR]

Most existing feature selection methods for object tracking assume that the samples in the previous frames are governed by the same distribution of the labeled samples obtained in the current frame and unlabeled samples collected in the next frame. However, according to our statistical analysis on very common videos, this assumption is not true in many scenarios. As a result, the selected features are not suitable for discriminating between the target from the background in the next frame. A tracking error accumulates and finally the drift problem happens. In this paper, we consider data distribution in tracking from a new perspective to adapt to target's and background's changes. We classify the samples into three categories: auxiliary samples (samples in the previous frames), target samples (samples collected in the current frame), and unlabeled samples (samples obtained in the next frame). To make the best use of them for tracking, we propose a novel semisupervised transfer learning approach that treats samples differently. Specifically, we assume that only target samples follow the same distribution as the unlabeled samples that we want to classify. Then, a novel and interesting semisupervised CovBoost method is developed utilizing the information provided by the three kinds of samples effectively when training the best strong classifier for tracking. Furthermore, we develop a new online updating algorithm for semisupervised CovBoost, making our tracker handle with significant variations of the tracked target and background successfully. Our experimental results demonstrate the advantages of treating samples differently during tracking. Our tracker outperforms state-of-the-art trackers on the benchmark datasets. [ABSTRACT FROM PUBLISHER]

Abstract: Currently, Nearest-Neighbor approaches (NN) have been applied to large scale real world image data mining. However, the following three disadvantages prevent them from wider application compared to other machine learning methods: (i) the performance is inferior on small datasets; (ii) the performance will degrade for data with high dimensions; (iii) they are heavily dependent on the chosen feature and distance measure. In this paper, we try to overcome the three mentioned intrinsic weaknesses by taking the abundant and diversified content of social media images into account. Firstly, we propose a novel neighborhood similarity measure which encodes both the local density information and semantic information, thus it has better generalization power than the original image-to-image similarity. Secondly, to enhance the scalability, we adopt kernelized Locality Sensitive Hashing (KLSH) to conduct approximated nearest neighbor search by utilizing a set of kernels calculated on several complementary image features. Finally, to enhance the robustness on diversified genres of images, we propose to fuse the discrimination power of different features by combining multiple neighborhood similarities calculated on different features/kernels with the entire retrieved nearest labeled and unlabeled image via the hashing systems. Experimental results on visual categorization on the Caltech-256 and two social media databases show the advantage of our method over traditional NN methods using the labeled data only. [Copyright &y& Elsevier]

As a successor to the light‐emitting‐diode (LED), the laser diode (LD) forecasts a brighter future for energy‐saving solid‐state lighting (SSL). Aiming for high‐performance white LD, herein, a brand‐new material form for Y3Al5O12:Ce3+ (YAG:Ce) garnet, that is, the integrated composite of YAG:Ce phosphor‐in‐silica glass ceramic/phosphor‐in‐silica ceramic (PiSGC/PiSC) film‐on‐sapphire plate (SP), following an "amorphous → polycrystalline" phase transformation mechanism, is developed. Remarkably, nearly full crystallization from glass in a controllable manner can be achieved. The in situ precipitated SiO2 microcrystals with high thermal conductivity (TC) favor to build efficient local heat dissipation network inside this opto‐functional composite, and therefore, substantially increase luminescence saturation threshold and luminous brightness under blue LD irradiation. Upon material optimization, bright and directional white light is generated with luminous flux (LF) of 1350 lm@10.5 W mm−2, correlated color temperature (CCT) of 7916 K, and color rendering index (CRI) of 69.3, enabling the application in high‐brightness long‐distance illumination scenarios. This smart material design highlights the combined merits of admirable luminescent properties from YAG:Ce@SiO2 ceramic, ultrahigh TC from sapphire single crystal, as well as, low cost, facile preparation, scalable manufacturing inherited from glass. [ABSTRACT FROM AUTHOR]

In this paper, we propose an attention-based virtual content insertion solution, called @ICT. Virtual content insertion (VCI) is an emerging application of video analysis and has been used in video augmentation and advertisement insertion. An ideal VCI solution should make the inserted virtual content being noticed by audiences and at the same time should not interfere with audiences' viewing experience on the original content. To balance these two conflicting issues, meaning high attention and low intrusiveness, we choose higher attentive shots as insertion time while determine insertion place and content interdependently by considering lower attention together with visual consistency. We also propose a measurement of intrusiveness from the viewpoint of visual attention. Furthermore, @ICT includes an in-scene insertion module, which embeds the virtual content into the videos with higher vividness and lower intrusiveness. @ICT is able to obtain an optimal balance between the noticing of the virtual content by audiences and disruption of viewing experience to the original content. It needs little prior knowledge and is applied to general videos. Extensive quantitative and qualitative evaluations on the VCI result have verified the effectiveness of the solution. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, we propose a new feature subset evaluation method for feature selection in object tracking. According to the fact that a feature which is useless by itself could become a good one when it is used together with some other features, we propose to evaluate feature subsets as a whole for object tracking instead of scoring each feature individually and find out the most distinguishable subset for tracking. In the paper, we use a special tree to formalize the feature subset space. Then conditional entropy is used to evaluating feature subset and a simple but efficient greedy search algorithm is developed to search this tree to obtain the optimal k-feature subset quickly. Furthermore, our online k-feature subset selection method is integrated into particle filter for robust tracking. Extensive experiments demonstrate that k-feature subset selected by our method is more discriminative and thus can improve tracking performance considerably. [Copyright &y& Elsevier]

This paper presents a novel rate control technique for multiview video plus depth (MVD) based 3D video coding. In the proposed rate control technique, an image-stitching method is first utilized to simultaneously encode video and depth, and then a joint rate control algorithm for MVD is presented. The joint rate control algorithm is performed on three levels, namely view level, video/depth level and frame level. In the view level, different proportions of rates are allocated for different types of views according to the pre-statistical rate allocation. In the video/depth level, the target rates for video and depth are discriminatorily assigned to guarantee the high quality of video for the backward-compatible display. In the frame level, the hierarchical rate allocation is used to regulate the target bits for each frame. In addition to the above mentioned rate control strategies, according to the special characteristics of multiview hypothetical reference decoder (HRD), the buffer-related rate control is also considered to prevent the decoder buffer from overflow or underflow even outputting multiple views. Experimental results show that the proposed rate control technique can accurately control the bit-rate to satisfy the requirements of 3D video systems. [ABSTRACT FROM AUTHOR]

In object detection, disparities in distributions between the training samples and the test ones are often inevitable, resulting in degraded performance for application scenarios. In this paper, we focus on the disparities caused by viewpoint and scene changes and propose an efficient solution to these particular cases by adapting generic detectors, assuming boosting style. A pretrained boosting-style detector encodes a priori knowledge in the form of selected features and weak classifier weighting. Towards adaptiveness, the selected features are shifted to the most discriminative locations and scales to compensate for the possible appearance variations. Moreover, the weighting coefficients are further adapted with covariate boost, which maximally utilizes the related training data to enrich the limited new examples. Extensive experiments validate the proposed adaptation mechanism towards viewpoint and scene adaptiveness and show encouraging improvement on detection accuracy over state-of-the-art methods. [ABSTRACT FROM PUBLISHER]

Abstract: This paper presents a rate-distortion (RD) optimized interactive streaming method for multiview video pre-compressed by H.264 Joint Multiview Video Model (JMVM). In the proposed method, multiple encodings are first used to facilitate the flexible server–client interaction. Second, a RD-optimized scheduling strategy is provided to guarantee the optimal view-dependent delivery of multiview video. In the RD-optimized scheduling strategy, a distortion model is proposed to estimate the expected end-to-end distortion by accounting for both coding and packet-loss-induced distortions, as well as rendering-induced distortion. With the end-to-end distortion model, the server can select the optimal encoding combination for transmission. Experimental results demonstrate that the proposed method can achieve a significant end-to-end RD performance improvement over the selective streaming methods with simulcast coding or scalable multiview coding. In addition, it has better error-resilience performance to combat with packet-losses over the Internet protocol (IP) networks. [Copyright &y& Elsevier]

Abstract: Joint video/depth rate allocation is an important optimization problem in 3D video coding. To address this problem, this paper proposes a distortion model to evaluate the synthesized view without access to the captured original view. The proposed distortion model is an additive model that accounts for the video-coding-induced distortion and the depth-quantization-induced distortion, as well as the inherent geometry distortion. Depth-quantization-induced distortion not only considers the warping error distortion, which is described by a piecewise linear model with the video power spectral property, but also takes into account the warping error correlation distortion between two sources reference views. Geometry distortion is approximated from that of the adjacent view synthesis. Based on the proposed distortion model, a joint rate allocation method is proposed to seek the optimal trade-off between video bit-rate and depth bit-rate for maximizing the view synthesis quality. Experimental results show that the proposed distortion model is capable of approximately estimating the actual distortion for the synthesized view, and that the proposed rate allocation method can almost achieve the identical rate allocation performance as the full-search method at less computational cost. Moreover, the proposed rate allocation method consumes less computational cost than the hierarchical-search method at high bit-rates while providing almost the equivalent rate allocation performance. [Copyright &y& Elsevier]

Abstract: Traditional Chinese painting (TCP) is the gem of Chinese traditional arts. More and more TCP images are digitized and exhibited on the Internet. Effectively browsing and retrieving them is an important problem that needs to be addressed. Gongbi (traditional Chinese realistic painting) and Xieyi (freehand style) are two basic types of traditional Chinese paintings. This paper proposes a scheme to detect TCPs from general images and categorize them into Gongbi and Xieyi schools. Low-level features such as color histogram, color coherence vectors, autocorrelation texture features and the newly proposed edge-size histogram are used to achieve the high-level classification. Support vector machine (SVM) is applied as the main classifier to obtain satisfactory classification results. Experimental results show the effectiveness of the method. [Copyright &y& Elsevier]

Abstract: Text in images and video frames carries important information for visual content understanding and retrieval. In this paper, by using multiscale wavelet features, we propose a novel coarse-to-fine algorithm that is able to locate text lines even under complex background. First, in the coarse detection, after the wavelet energy feature is calculated to locate all possible text pixels, a density-based region growing method is developed to connect these pixels into regions which are further separated into candidate text lines by structural information. Secondly, in the fine detection, with four kinds of texture features extracted to represent the texture pattern of a text line, a forward search algorithm is applied to select the most effective features. Finally, an SVM classifier is used to identify true text from the candidates based on the selected features. Experimental results show that this approach can fast and robustly detect text lines under various conditions. [Copyright &y& Elsevier]

Abstract: By adaptively selecting image window size based on the pyramid data structure manipulation of Lorentz information measure, a new technique for image thresholding is proposed. The advantage of this technique is its effectiveness in eliminating both uneven lighting disturbance and ghost objects. When applied to Otsu’s thresholding approach, it can provide accurate result under uneven lighting disturbance while using Otsu’s method alone cannot. Experimental results show the effectiveness of this method. [Copyright &y& Elsevier]

A 9-10-bit adjustable and energy-efficient switching scheme for SAR ADC with one-LSB common-mode voltage variation is proposed. Based on capacitor-splitting technology and common-mode conversion techniques, the proposed switching scheme reduces the DAC switching energy by 96.41% compared to the conventional scheme. The low complexity and the one-LSB common-mode voltage offset of this scheme benefit from the simultaneous switching of the reference voltages of the capacitors corresponding to the positive array and the negative array throughout the entire reference voltage switching process, and the reference voltage of each capacitor in the scheme does not change more than two voltages. The post-layout result shows that the ADC achieves the 54.96 dB SNDR, the 61.73 dB SFDR, and the 0.67 μw power consumption with the 10-bit mode and the 48.33 dB SNDR, the 54.17 dB SFDR, and the 0.47 μw power consumption with the 9-bit mode in a 180 nm process with a 100 kS/s sampling frequency. [ABSTRACT FROM AUTHOR]

Image quality assessment (IQA) plays a central role in many image processing algorithms and systems. Although many popular IQA models achieves high performance on existing released databases, they are still not well accepted in practical applications due to the not-always satisfactory accuracy on real-world data and situations. In this paper, we revisit the IQA research, and point out an ignored but interesting problem in IQA: the coarse-grained (i.e., when quality variation is sufficiently big, as the setting of most IQA databases up to date) statistical results evaluated on existing databases mask the fine-grained differentiation. Accordingly, we present a survey on image quality assessment from a new perspective: fine-grained image quality assessment (FG-IQA). Recent FG-IQA research on five major kinds of images is introduced, and some popular IQA methods are analyzed from FG-IQA perspective. The potential problems for current IQA research based on existing coarse-grained databases are analyzed and the necessity of more FG-IQA research is justified. Finally, we discuss some challenges and possible directions for future works in FG-IQA. [ABSTRACT FROM AUTHOR]

The preparation of high performance upconversion luminescence materials with stronger shortwave emissions is still a challenge. In this work, the feature of hexagon circles sublattice structure of tetragonal LiYF 4 was revealed, and the sublattice level energy clusters of 4 Yb3+-1Tm3+ were constructed by the substitution of 70 mol% Yb3+ for Y3+ in tetragonal LiYF 4 matrix. The sensitizing ions concentration quenching was inhibited by the hexagon circles sublattice structure, and the upconversion luminescence was dramatically enhanced, especially, the UV band emission was enhanced about 20 times and the intensity ratio of UV to blue emissions was improved about 10 folds under the excitation power density of 0.230 W/cm2. Photon numbers of upconversion emissions obviously decreased with the increasing of sublattice level energy clusters that the upconversion process of three photons was changed to two steps of two photons process. The upconversion luminescence mechanism of sublattice energy cluster construction was system investigated. The material with high performance of UV and blue UC luminescence can be a well candidate for the applications of photodynamic therapy and photocatalytic, etc. The work may be an inspiration for high efficient UC luminescence materials design and preparation. • The feature of hexagon circles sublattice structure of tetragonal LiYF 4 was firstly revealed. • The UV band emission was dramatically enhanced about 20 times by the construction of sublattice level energy clusters of 4 Yb3+-1Tm3+. • The sensitizing ions concentration quenching was effectively inhibited by the hexagon circles sublattice structure. • Three Photons upconversion process was changed to two steps of two photons upconversion process by the construction of sublattice level energy clusters. [ABSTRACT FROM AUTHOR]

Recently, fully convolutional networks (FCNs) have made great progress in the task of salient object detection and existing state-of-the-arts methods mainly focus on how to integrate edge information in deep aggregation models. In this paper, we propose a novel Decomposition and Completion Network (DCN), which integrates edge and skeleton as complementary information and models the integrity of salient objects in two stages. In the decomposition network, we propose a cross multi-branch decoder, which iteratively takes advantage of cross-task aggregation and cross-layer aggregation to integrate multi-level multi-task features and predict saliency, edge, and skeleton maps simultaneously. In the completion network, edge and skeleton maps are further utilized to fill flaws and suppress noises in saliency maps via hierarchical structure-aware feature learning and multi-scale feature completion. Through jointly learning with edge and skeleton information for localizing boundaries and interiors of salient objects respectively, the proposed network generates precise saliency maps with uniformly and completely segmented salient objects. Experiments conducted on five benchmark datasets demonstrate that the proposed model outperforms existing networks. Furthermore, we extend the proposed model to the task of RGB-D salient object detection, and it also achieves state-of-the-art performance. The code is available at https://github.com/wuzhe71/DCN. [ABSTRACT FROM AUTHOR]

This paper studies vehicle ReID in aerial videos taken by Unmanned Aerial Vehicles (UAVs). Compared with existing vehicle ReID tasks performed with fixed surveillance cameras, UAV vehicle ReID is still under-explored and could be more challenging, e.g., aerial videos have dynamic and complex backgrounds, different vehicles show similar appearance, and the same vehicle commonly show distinct viewpoints and scales. To facilitate the research on UAV vehicle ReID, this paper contributes a novel dataset called UAV-VeID. UAV-VeID contains 41,917 images of 4601 vehicles captured by UAVs, where each vehicle has multiple images taken from different viewpoints. UAV-VeID also includes a large-scale distractor set to encourage the research on efficient ReID schemes. Compared with existing vehicle ReID datasets, UAV-VeID exhibits substantial variances in viewpoints and scales of vehicles, thus requires more robust features. To alleviate the negative effects of those variances, this paper also proposes a viewpoint adversarial training strategy and a multi-scale consensus loss to promote the robustness and discriminative power of learned deep features. Extensive experiments on UAV-VeID show our approach outperforms recent vehicle ReID algorithms. Moreover, our method also achieves competitive performance compared with recent works on existing vehicle ReID datasets including VehicleID, VeRi-776 and VERI-Wild. [ABSTRACT FROM AUTHOR]

Vehicle Re-Identification (Re-ID) is a challenging vision task mainly because the appearance of a vehicle varies dramatically under different viewpoints. Moreover, different vehicles with the same model and color commonly show similar appearance, thus are hard to be distinguished. To alleviate negative effects of viewpoint variance, we design a multi-view branch network where each branch learns a viewpoint-specific feature without parameter sharing. Being able to focus on a limited range of viewpoints, this viewpoint-specific feature performs substantially better than the general feature learned by an uniform network. To further differentiate visually similar vehicles, we strengthen the discriminative power on their subtle local differences by introducing a spatial attention model into each feature learning branch. The multi-view feature learning and spatial attention learning compose our neural network architecture, which is trained end to end with the softmax loss and triplet loss, respectively. We evaluate our methods on two large vehicle Re-ID datasets, i.e., VehicleID and VeRi-776, respectively. Extensive experiments show that our methods achieve promising performance. For example, we achieve mAP accuracy of 76.78% and 72.53% on VehicleID and VeRi-776 dataset respectively, substantially better than current state-of-the art. [ABSTRACT FROM AUTHOR]

• SiO 2 -TiO 2 composite powders are constructed under milder conditions. • Submicron-sized Ti-rich MWW is prepared via mechanochemical dry gel conversion (MDGC). • Ti-MWW-MDGC improves catalytic activity remarkably than Ti-MWW-DGC. • MDGC route is a promising approach to green synthesis of highly active zeolites. In this research, a novel green mechanochemical dry gel conversion method (MDGC) is reported to obtain submicron-sized Ti-rich MWW zeolite. First, titanium and silica sources were milled under milder conditions in a planetary ball mill to construct the uniform SiO 2 -TiO 2 composite powders with higher Ti contents. Subsequently, submicron-sized Ti-rich MWW zeolite powder was prepared via dry gel conversion (DGC) route using the obtained composite under mixed steam environment consisting of water and piperidine. The physicochemical properties of the SiO 2 -TiO 2 composites before and after calcination and corresponding Ti-MWW-MDGC were investigated by XRD, UV–Vis, FT-IR, SEM, XRF, and N 2 absorption-desorption. Moreover, Ti-MWW (30)-MDGC exhibited a superior catalytic activity during epoxidation with H 2 O 2 compared with the traditional DGC method, and the conversion of 1-hexene increased from 4.2% to 26.3%. The MDGC method is a promising approach for the green production of various zeolites with higher heteroatom contents, such as TS-1, Ti-Beta and Sn-MWW, etc. [ABSTRACT FROM AUTHOR]

Visual saliency provides important information for understanding scenes in many computer vision tasks. The existing video saliency algorithms mainly focus on predicting spatial and temporal saliency maps. However, these maps are simply fused without considering the complex dynamic scenes in videos. To overcome this drawback, we propose a deep convolutional fusion framework for video saliency prediction. The proposed model, which is based on coupled fully convolutional networks (FCNs), effectively encodes the spatiotemporal information by integrating spatial and temporal features. We demonstrate that this information is helpful for accurately fusing the spatial and temporal saliency maps according to changes in video scenes. In particular, we gradually design three different deep fusion architectures to investigate how to better utilize the spatiotemporal information. Moreover, we propose a reasonable sampling strategy for selecting suitable training sets for the coupled FCNs. Through extensive experiments, we demonstrate that our model outperforms the state-of-the-art algorithms on four public video saliency data sets. [ABSTRACT FROM AUTHOR]

Abstract Cross-modal retrieval has gained much attention in recent years. As the research mainstream, most of existing approaches learn projections for data from different modalities into a common space where data can be compared directly. However, they neglect the preservation of feature and semantic information, so they are unable to obtain satisfactory results as expected. In this paper, we propose a two-stage learning method to learn multi-modal mappings that project multi-modal data to low dimensional embeddings that preserve both feature and semantic information. In the first stage, we combine both low-level feature and high-level semantic information to learn feature-aware semantic code vectors. In the second stage, we use encoder–decoder paradigm to learn projections. The encoder projects feature vectors to code vectors, and the decoder projects code vectors back to feature vectors. The encoder-decoder paradigm guarantees the embeddings to preserve both feature and semantic information. An alternating minimization procedure is developed to solve the multi-modal semantic autoencoder optimization problem. Extensive experiments on three benchmark datasets demonstrate that the proposed method outperforms state-of-the-art cross-modal retrieval methods. [ABSTRACT FROM AUTHOR]

Sustainable synthesis of MWW zeolites have been the focus of attention but still a significant challenge. In this study, an innovative synthesis strategy of the EniRicerche-Boralite-1 (ERB-1) precursor is reported via an interesting steam-environment crystallization method (SEC). Compared with the conventional hydrothermal synthesis, this method not only simplified the synthetic process but also reduced 85% amount of boric acid and solvent and half amount of the piperidine (PI, SDA, structure-directing agent). Moreover, the crystallization time was also reduced from 7 d to 2 d. The Ti-MWW prepared with this ERB-1 precursor showed the same catalytic performance as typical post-synthesized Ti-MWW (Ti-MWW-PS) in the liquid-phase epoxidation of 1-hexene with H 2 O 2 . XRD and SEM techniques explained the effects of boron content and crystallization time on the preparation of ERB-1. The physicochemical properties of the ERB-1 and corresponding Ti-MWW in different preparation conditions were investigated systematically by the FT-IR, UV–vis, ICP, NH 3 -TPD and N 2 absorption-desorption techniques. This method is both eco-friendly for the synthesis of ERB-1 and effective for the preparation of MWW-type zeolites in industrial production on a large scale in the future. [ABSTRACT FROM AUTHOR]

Graph convolution networks (GCNs) based methods for 3D human pose estimation usually aggregate immediate features of single-hop nodes, which are unaware of the correlation of multi-hop nodes and therefore neglect long-range dependency for predicting complex poses. In addition, they typically operate either on single-scale or sequential down-sampled multi-scale graph representations, resulting in the loss of contextual information or spatial details. To address these problems, this paper proposes a parallel hop-aware graph attention network (PHGANet) for 3D human pose estimation, which learns enriched hop-aware correlation of the skeleton joints while maintaining the spatially-precise representations of the human graph. Specifically, we propose a hop-aware skeletal graph attention (HSGAT) module to capture the semantic correlation of multi-hop nodes, which first calculates skeleton-based 1-hop attention and then disseminates it to arbitrary hops via graph connectivity. To alleviate the redundant noise introduced by the interactions with distant nodes, HSGAT uses an attenuation strategy to separate attention from distinct hops and assign them learnable attenuation weights according to their distances adaptively. Upon HSGAT, we further build PHGANet with multiple parallel branches of stacked HSGAT modules to learn the enriched hop-aware correlation of human skeletal structures at different scales. In addition, a joint centrality encoding scheme is proposed to introduce node importance as a bias in the learned graph representation, which makes the core joints (e.g., neck and pelvis) more influential during node aggregation. Experimental results indicate that PHGANet performs favorably against state-of-the-art methods on the Human3.6M and MPI-INF-3DHP benchmarks. Models and code are available at https://github.com/ChenyangWang95/PHGANet/. [ABSTRACT FROM AUTHOR]

Multiple object tracking is a key problem for many computer vision applications such as video surveillance, advanced driver assistance or animation. Most of existing tracking-by-detection methods are mainly based on object appearances and motions. However, the contextual information around the target has not been fully exploited. In this paper, we pay more attention to the contextual information and propose an Exchanging Object Context (EOC) model, which takes full advantage of the context information. Specifically, we implement an efficient and accurate online multiple object tracking algorithm with a novel affinity measure to associate detections. This measure calculates the similarity between targets and detections with the background smoothness after exchanging the contexts between detections and targets, using a novel color histogram descriptor. We refine the bounding boxes by measuring the context changes. Extensive experimental results on two public benchmarks demonstrate the effectiveness of the proposed tracking method with comparisons to several state-of-the-art trackers. [ABSTRACT FROM AUTHOR]

The advent of wide color-gamut LED backlight technology greatly improves the user experience in visions. The bottleneck restricting progress of this technology is to develop red phosphor with satisfactory spectral features and low cost. Herein, a novel non-rare-earth K2XF7:Mn4+ (X = Ta, Nb) was successfully designed and synthesized, exhibiting an admirable quantum efficiency of ∼93.5%, an extremely narrow FWHM of ∼2.3 nm, and a high color purity of ∼99.6%. Unlike the previously reported Mn4+ activated fluorides, Mn4+ in K2XF7 suffers highly distorted octahedral environment in C1 (or C2v) group symmetry, enabling the appearance of intense zero phonon line and distinctive vibronic transitions. Systematical high-resolution spectroscopic analyses down to 10 K disclose the influences of strong crystal field, weak nephelauxetic effect, and intermediate electron-phonon coupling are responsible for the peculiar spectral features of Mn4+ in K2XF7. Impressively, a wide color-gamut reaching up to 86.7% NTSC is realized, demonstrating great validity of K2XF7:Mn4+ for LCD backlights. The present study not only brings a brand-new kind of Mn4+ activated host to the sight of phosphor community (the host can be extended to surges of tantalum/niobium fluorides), but also may enlighten researchers to design highly distorted environment to achieve unique spectral properties of Mn4+. [ABSTRACT FROM AUTHOR]

Visual features are unsatisfactory to effectively describe the visual semantics. However, single layer based semantic modeling may be not able to cope with complicated semantic contents. In this paper, we propose Hierarchical Deep Semantic Representation (H-DSR), a hierarchical framework which combines semantic context modeling with visual features. First, the input image is sampled with spatially fixed grids. Deep features are then extracted for each sample in particular location. Second, using pre-learned classifiers, a detection response map is constructed for each patch. Semantic representation is then extracted from the map, which have a sense of latent semantic context. We combine the semantic and visual representations for joint representation. Third, a hierarchical deep semantic representation is built with recurrent reconstructions using three layers. The concatenated visual and semantic representations are used as the inputs of subsequent layers for semantic representation extraction. Finally, we verify the effectiveness of H-DSR for visual categorization on two publicly available datasets: Oxford Flowers 17 and UIUC-Sports. Improved performances are obtained over many baseline methods. [ABSTRACT FROM AUTHOR]

Data from real applications involve multiple modalities representing content with the same semantics from complementary aspects. However, relations among heterogeneous modalities are simply treated as observation-to-fit by existing work, and the parameterized modality specific mapping functions lack flexibility in directly adapting to the content divergence and semantic complicacy in multimodal data. In this paper, we build our work based on the Gaussian process latent variable model (GPLVM) to learn the non-parametric mapping functions and transform heterogeneous modalities into a shared latent space. We propose multimodal Similarity Gaussian Process latent variable model (m-SimGP), which learns the mapping functions between the intra-modal similarities and latent representation. We further propose multimodal distance-preserved similarity GPLVM (m-DSimGP) to preserve the intra-modal global similarity structure, and multimodal regularized similarity GPLVM (m-RSimGP) by encouraging similar/dissimilar points to be similar/dissimilar in the latent space. We propose m-DRSimGP, which combines the distance preservation in m-DSimGP and semantic preservation in m-RSimGP to learn the latent representation. The overall objective functions of the four models are solved by simple and scalable gradient decent techniques. They can be applied to various tasks to discover the nonlinear correlations and to obtain the comparable low-dimensional representation for heterogeneous modalities. On five widely used real-world data sets, our approaches outperform existing models on cross-modal content retrieval and multimodal classification. [ABSTRACT FROM AUTHOR]

Binary Relevance is a well-known framework for multi-label classification, which considers each class label as a binary classification problem. Many existing multi-label algorithms are constructed within this framework, and utilize identical data representation in the discrimination of all the class labels. In multi-label classification, however, each class label might be determined by some specific characteristics of its own. In this paper, we seek to learn label-specific data representation for each class label, which is composed of label-specific features. Our proposed method LLSF can not only be utilized for multi-label classification directly, but also be applied as a feature selection method for multi-label learning and a general strategy to improve multi-label classification algorithms comprising a number of binary classifiers. Inspired by the research works on modeling high-order label correlations, we further extend LLSF to learn class-Dependent Labels in a sparse stacking way, denoted as LLSF-DL. It incorporates both second-order and high-order label correlations. A comparative study with the state-of-the-art approaches manifests the effectiveness and efficiency of our proposed methods. [ABSTRACT FROM PUBLISHER]

Traffic control is an effective and efficient method for the problem of traffic congestion. It is necessary to design a high-level controller to regulate the network traffic demands, because traffic congestion is not only caused by the improper management of the traffic network but also to a great extent caused by excessive network traffic demands. Therefore, we design a demand-balance model predictive controller based on the macroscopic fundamental diagram-based multi-subnetwork model, which can optimize the network traffic mobility and the network traffic throughput by regulating the input traffic flows of the subnetworks. Because the transferring traffic flows among subnetworks are indirectly controlled and coordinated by the demand-balance model predictive controller, the subnetwork division can variate dynamically according to real traffic states, and a global optimality can be achieved for the entire traffic network. The simulation results show the effectiveness of the proposed controller in improving the network traffic throughput. Copyright © 2016 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Click-through rate (CTR) prediction, whose goal is to predict the probability of the user to click on an item, has become increasingly significant in the recommender systems. Recently, some deep learning models with the ability to automatically extract the user interest from his/her behaviors have achieved great success. In these work, the attention mechanism is used to select the user interested items in historical behaviors, improving the performance of the CTR predictor. Normally, these attentive modules can be jointly trained with the base predictor by using gradient descents. In this paper, we regard user interest modeling as a feature selection problem, which we call user interest selection. For such a problem, we propose a novel approach under the framework of the wrapper method, which is named Meta-Wrapper. More specifically, we use a differentiable module as our wrapping operator and then recast its learning problem as a continuous bilevel optimization. Moreover, we use a meta-learning algorithm to solve the optimization and theoretically prove its convergence. Meanwhile, we also provide theoretical analysis to show that our proposed method 1) efficiencies the wrapper-based feature selection, and 2) achieves better resistance to overfitting. Finally, extensive experiments on three public datasets manifest the superiority of our method in boosting the performance of CTR prediction. [ABSTRACT FROM AUTHOR]

The Area under the ROC curve (AUC) is a well-known ranking metric for problems such as imbalanced learning and recommender systems. The vast majority of existing AUC-optimization-based machine learning methods only focus on binary-class cases, while leaving the multiclass cases unconsidered. In this paper, we start an early trial to consider the problem of learning multiclass scoring functions via optimizing multiclass AUC metrics. Our foundation is based on the M metric, which is a well-known multiclass extension of AUC. We first pay a revisit to this metric, showing that it could eliminate the imbalance issue from the minority class pairs. Motivated by this, we propose an empirical surrogate risk minimization framework to approximately optimize the M metric. Theoretically, we show that: (i) optimizing most of the popular differentiable surrogate losses suffices to reach the Bayes optimal scoring function asymptotically; (ii) the training framework enjoys an imbalance-aware generalization error bound, which pays more attention to the bottleneck samples of minority classes compared with the traditional $O(\sqrt{1/N})$ O (1 / N) result. Practically, to deal with the low scalability of the computational operations, we propose acceleration methods for three popular surrogate loss functions, including the exponential loss, squared loss, and hinge loss, to speed up loss and gradient evaluations. Finally, experimental results on 11 real-world datasets demonstrate the effectiveness of our proposed framework. The code is now available at https://github.com/joshuaas/Learning-with-Multiclass-AUC-Theory-and-Algorithms. [ABSTRACT FROM AUTHOR]

As pairwise ranking becomes broadly employed for elections, sports competitions, recommendation, information retrieval and so on, attackers have strong motivation and incentives to manipulate or disrupt the ranking list. They could inject malicious comparisons into the training data to fool the target ranking algorithm. Such a technique is called “poisoning attack” in regression and classification tasks. In this paper, to the best of our knowledge, we initiate the first systematic investigation of data poisoning attack on the pairwise ranking algorithms, which can be generally formalized as the dynamic and static games between the ranker and the attacker, and can be modeled as certain kinds of integer programming problems mathematically. To break the computational hurdle of the underlying integer programming problems, we reformulate them into the distributionally robust optimization (DRO) problems, which are computational tractable. Based on such DRO formulations, we propose two efficient poisoning attack algorithms and establish the associated theoretical guarantees including the existence of Nash equilibrium and the generalization ability bounds. The effectiveness of the suggested poisoning attack strategies is demonstrated by a series of toy simulations and several real data experiments. These experimental results show that the proposed methods can significantly reduce the performance of the ranker in the sense that the correlation between the true ranking list and the aggregated results with toxic data can be decreased dramatically. [ABSTRACT FROM AUTHOR]

Focusing on the issue of how to effectively capture and utilize cross-modality information in RGB-D salient object detection (SOD) task, we present a convolutional neural network (CNN) model, named CIR-Net, based on the novel cross-modality interaction and refinement. For the cross-modality interaction, 1) a progressive attention guided integration unit is proposed to sufficiently integrate RGB-D feature representations in the encoder stage, and 2) a convergence aggregation structure is proposed, which flows the RGB and depth decoding features into the corresponding RGB-D decoding streams via an importance gated fusion unit in the decoder stage. For the cross-modality refinement, we insert a refinement middleware structure between the encoder and the decoder, in which the RGB, depth, and RGB-D encoder features are further refined by successively using a self-modality attention refinement unit and a cross-modality weighting refinement unit. At last, with the gradually refined features, we predict the saliency map in the decoder stage. Extensive experiments on six popular RGB-D SOD benchmarks demonstrate that our network outperforms the state-of-the-art saliency detectors both qualitatively and quantitatively. The code and results can be found from the link of https://rmcong.github.io/proj_CIRNet.html. [ABSTRACT FROM AUTHOR]

Generation of a 3D model of an object from multiple views has a wide range of applications. Different parts of an object would be accurately captured by a particular view or a subset of views in the case of multiple views. In this paper, a novel coarse-to-fine network (C2FNet) is proposed for 3D point cloud generation from multiple views. C2FNet generates subsets of 3D points that are best captured by individual views with the support of other views in a coarse-to-fine way, and then fuses these subsets of 3D points to a whole point cloud. It consists of a coarse generation module where coarse point clouds are constructed from multiple views by exploring the cross-view spatial relations, and a fine generation module where the coarse point cloud features are refined under the guidance of global consistency in appearance and context. Extensive experiments on the benchmark datasets have demonstrated that the proposed method outperforms the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Transferable adversarial attacks against Deep neural networks (DNNs) have received broad attention in recent years. An adversarial example can be crafted by a surrogate model and then attack the unknown target model successfully, which brings a severe threat to DNNs. The exact underlying reasons for the transferability are still not completely understood. Previous work mostly explores the causes from the model perspective, e.g., decision boundary, model architecture, and model capacity. Here, we investigate the transferability from the data distribution perspective and hypothesize that pushing the image away from its original distribution can enhance the adversarial transferability. To be specific, moving the image out of its original distribution makes different models hardly classify the image correctly, which benefits the untargeted attack, and dragging the image into the target distribution misleads the models to classify the image as the target class, which benefits the targeted attack. Towards this end, we propose a novel method that crafts adversarial examples by manipulating the distribution of the image. We conduct comprehensive transferable attacks against multiple DNNs to demonstrate the effectiveness of the proposed method. Our method can significantly improve the transferability of the crafted attacks and achieves state-of-the-art performance in both untargeted and targeted scenarios, surpassing the previous best method by up to 40% in some cases. In summary, our work provides new insight into studying adversarial transferability and provides a strong counterpart for future research on adversarial defense. [ABSTRACT FROM AUTHOR]

Video captioning aims to generate a natural language sentence to describe the main content of a video. Since there are multiple objects in videos, taking full exploration of the spatial and temporal relationships among them is crucial for this task. The previous methods wrap the detected objects as input sequences, and leverage vanilla self-attention or graph neural network to reason about visual relations. This cannot make full use of the spatial and temporal nature of a video, and suffers from the problems of redundant connections, over-smoothing, and relation ambiguity. In order to address the above problems, in this paper we construct a long short-term graph (LSTG) that simultaneously captures short-term spatial semantic relations and long-term transformation dependencies. Further, to perform relational reasoning over the LSTG, we design a global gated graph reasoning module (G3RM), which introduces a global gating based on global context to control information propagation between objects and alleviate relation ambiguity. Finally, by introducing G3RM into Transformer instead of self-attention, we propose the long short-term relation transformer (LSRT) to fully mine objects’ relations for caption generation. Experiments on MSVD and MSR-VTT datasets show that the LSRT achieves superior performance compared with state-of-the-art methods. The visualization results indicate that our method alleviates problem of over-smoothing and strengthens the ability of relational reasoning. [ABSTRACT FROM AUTHOR]

Most of the previous works for multitarget tracking employ two strategies: global optimization and online state estimation. In general, global methods attempt to prevent local optimization and find the best results given global models. However, in time-critical applications, global optimization has long temporal latency. In contrast, most of the online algorithms obtain the states with greedy estimation, in which the errors are hard to be corrected. In this paper, we combine these two strategies and propose an online tracking approach with short-term storage to correct some local association errors. Based on structured output support vector machine, we propose a new framework with multiple online learners to produce multiple best local linkages, and novel features based on previously generated multiframe associations are designed for reranking of these multiple linkages. The reranking also serves as the appropriate mediator for updating of the online learners by co-train algorithm. The experimental results illustrate the advantage and robustness of this reranking algorithm, and its discrimination to find optimal ones. Comparison with some state-of-the-art methods proves that our proposed method is competitive to global optimal ones and is superior to other online tracking algorithms. [ABSTRACT FROM PUBLISHER]

In this paper, we study the problem of estimating subjective visual properties (SVP) for images, which is an emerging task in Computer Vision. Generally speaking, collecting SVP datasets involves a crowdsourcing process where annotations are obtained from a wide range of online users. Since the process is done without quality control, SVP datasets are known to suffer from noise. This leads to the issue that not all samples are trustworthy. Facing this problem, we need to develop robust models for learning SVP from noisy crowdsourced annotations. In this paper, we construct two general robust learning frameworks for this application. Specifically, in the first framework, we propose a probabilistic framework to explicitly model the sparse unreliable patterns that exist in the dataset. It is noteworthy that we then provide an alternative framework that could reformulate the sparse unreliable patterns as a “contraction” operation over the original loss function. The latter framework leverages not only efficient end-to-end training but also rigorous theoretical analyses. To apply these frameworks, we further provide two models as implementations of the frameworks, where the sparse noise parameters could be interpreted with the HodgeRank theory. Finally, extensive theoretical and empirical studies show the effectiveness of our proposed framework. [ABSTRACT FROM AUTHOR]