Search

Learning discriminative features is of vital importance for automatic facial expression recognition (FER) in the wild. In this article, we propose a novel Slide-Patch and Whole-Face Attention model with SE blocks (SPWFA-SE), which jointly perceives the discriminative locality characteristics and informative global features of the face for effective FER. Specifically, the well-designed slide patches are proposed to extract local features. Different from the existing methods, our slide patches not only can maintain the information at the edge area of patches, but also do not need to detect facial landmarks. Moreover, to make the model adaptively focus on the distinguishable regions, an attention module is proposed in the patch level to learn the weight of each patch. Furthermore, squeeze-and-excitation blocks are explored in the channel level to learn the weight of each channel. As such, the proposed multi-level feature extraction and attention mechanisms can enhance the representative ability of the learned features. Extensive experiments on five challenging datasets demonstrate that our method can achieve state-of-the-art performance. Cross database experiments on another three databases show the superior generalization performance of our model. Furthermore, complexity analysis results show that our model contains fewer parameters with fast training advantages than other competing models. [ABSTRACT FROM AUTHOR]

In convolutional neural networks (CNNs), generating noise for the intermediate feature is a hot research topic in improving generalization. The existing methods usually regularize the CNNs by producing multiplicative noise (regularization weights), called multiplicative regularization (Multi-Reg). However, Multi-Reg methods usually focus on improving generalization but fail to jointly consider optimization, leading to unstable learning with slow convergence. Moreover, Multi-Reg methods are not flexible enough since the regularization weights are generated from a definite manual-design distribution. Besides, most popular methods are not universal enough, because these methods are only designed for the residual networks. In this article, we, for the first time, experimentally and theoretically explore the nature of generating noise in the intermediate features for popular CNNs. We demonstrate that injecting noise in the feature space can be transformed to generating noise in the input space, and these methods regularize the networks in a Mini-batch in Mini-batch (MiM) sampling manner. Based on these observations, this article further discovers that generating multiplicative noise can easily degenerate the optimization due to its high dependence on the intermediate feature. Based on these studies, we propose a novel additional regularization (Addi-Reg) method, which can adaptively produce additional noise with low dependence on intermediate feature in CNNs by employing a series of mechanisms. Particularly, these well-designed mechanisms can stabilize the learning process in training, and our Addi-Reg method can pertinently learn the noise distributions for every layer in CNNs. Extensive experiments demonstrate that the proposed Addi-Reg method is more flexible and universal, and meanwhile achieves better generalization performance with faster convergence against the state-of-the-art Multi-Reg methods. [ABSTRACT FROM AUTHOR]

The 360° image that offers a 360-degree scenario of the world is widely used in virtual reality and has drawn increasing attention. In 360° image compression, the spherical image is first transformed into a planar image with a projection such as equirectangular projection (ERP) and then saved with the existing codecs. The ERP images that represent different circles of latitude with the same number of pixels suffer from the unbalance sampling problem, resulting in inefficiency using planar compression methods, especially for the deep neural network (DNN) based codecs. To tackle this problem, we introduce a latitude adaptive coding scheme for DNNs by allocating variant numbers of codes for different regions according to the latitude on the sphere. Specifically, taking both the number of allocated codes for each region and their entropy into consideration, we introduce a flexible regional adaptive rate loss for region-wise rate controlling. Latitude adaptive constraints are then introduced to prevent spending too many codes on the over-sampling regions. Furthermore, we introduce viewport-based distortion loss by calculating the average distortion on a set of viewports. We optimize and test our model on a large 360° dataset containing 19,790 images collected from the Internet. The experiment results demonstrate the superiority of the proposed latitude adaptive coding scheme. On the whole, our model outperforms the existing image compression standards, including JPEG, JPEG2000, HEVC Intra Coding, and VVC Intra Coding, and helps to save around 15% bits compared to the baseline learned image compression model for planar images. [ABSTRACT FROM AUTHOR]

Single image super-resolution (SISR) using deep convolutional neural networks (CNNs) achieves the state-of-the-art performance. Most existing SISR models mainly focus on pursuing high peak signal-to-noise ratio (PSNR) and neglect textures and details. As a result, the recovered images are often perceptually unpleasant. To address this issue, in this paper, we propose a texture and detail-preserving network (TDPN), which focuses not only on local region feature recovery but also on preserving textures and details. Specifically, the high-resolution image is recovered from its corresponding low-resolution input in two branches. First, a multi-reception field based branch is designed to let the network fully learn local region features by adaptively selecting local region features in different reception fields. Then, a texture and detail-learning branch supervised by the textures and details decomposed from the ground-truth high resolution image is proposed to provide additional textures and details for the super-resolution process to improve the perceptual quality. Finally, we introduce a gradient loss into the SISR field and define a novel hybrid loss to strengthen boundary information recovery and to avoid overly smooth boundary in the final recovered high-resolution image caused by using only the MAE loss. More importantly, the proposed method is model-agnostic, which can be applied to most off-the-shelf SISR networks. The experimental results on public datasets demonstrate the superiority of our TDPN on most state-of-the-art SISR methods in PSNR, SSIM and perceptual quality. We will share our code on https://github.com/tocaiqing/TDPN. [ABSTRACT FROM AUTHOR]

Millimeter-wave (mmWave) massive multiple-input multiple-output (MIMO) has been considered as a key enabler for the fifth-generation (5G) communications. It is essential to design and test mmWave 5G devices under various realistic scenarios since the radio propagation channels pose intrinsic limitations on the performance. This requires emulating realistic dynamic mmWave channels in a reproducible manner in laboratories, which is the goal of this article. In this contribution, we first illustrate the dominant-cluster(s) concept, where the nondominant clusters in the mmWave channels are pruned, for mmWave 5G devices applying massive MIMO beamforming. This demonstrates the importance and necessity to accurately emulate the mmWave channels at a cluster level rather than the composite-channel level. Thus, an over-the-air (OTA) emulation strategy for dynamic mmWave channels is proposed based on the concept of the dominant cluster(s) in a sectored multiprobe anechoic chamber (SMPAC). The key design parameters, including the probe number and the angular spacing of probes, are investigated through comprehensive simulations. A cost-effective switch circuit is also designed for this purpose and validated in the simulation. Furthermore, a dynamic mmWave channel measured in an indoor scenario at 28–30 GHz is presented, where the proposed emulation strategy is also validated by reproducing the measured reality. [ABSTRACT FROM AUTHOR]

The informativeness and discriminativeness of features collaboratively ensure high-accuracy Facial Expression Recognition (FER) in the wild. Most of existing methods use the single-path deep convolutional neural network with softmax loss for basic FER, while they cannot deal with the challenging situations of the compound FER in the wild, because they fail to learn informative and discriminative features in a targeted manner. To this end, we present an Informative and Discriminative Feature Learning (IDFL) framework that consists of two key components: the Multi-Path Attention Convolutional Neural Network (MPACNN) and Balanced Separate loss (BS loss), for both basic and compound high-accuracy FER in the wild. Specifically, MPACNN leverages different paths to learn diverse features. These features are then adaptively fused into informative ones via an attention module, such that the model can adequately capture detailed information for both basic and compound FER. The BS loss maximizes the inter-class distance of features and minimizes the intra-class one. In this way, the features are discriminative enough for high-accuracy FER in the wild. Particularly, the BS loss is invoked as the objective function of MPACNN, so the model can learn informative and discriminative features at the same time, yielding better performance. Seven databases are utilized to evaluate the proposed method, and the results demonstrate that our method achieves state-of-the-art performance on both basic and compound expressions with good generalization ability. Moreover, our model contains fewer parameters and can be trained faster than other related models. [ABSTRACT FROM AUTHOR]

With the increased model size of convolutional neural networks (CNNs), overfitting has become the main bottleneck to further improve the performance of networks. Currently, the weighting regularization methods have been proposed to address the overfitting problem and they perform satisfactorily. Since these regularization methods cannot be used in all the networks and they are usually not flexible enough in different phases of the training and test processes, this article proposes a multiscale conditional (MSC) regularization method. MSC divides the intermediate features into different scales and then generates new data for each scale features, respectively. In addition, the new data are generated by employing the information from two conditions: 1) each sample feature and 2) each layer pattern. Finally, a self-identity structure is proposed to supplement the features with the generated data. Therefore, MSC can adaptively and efficiently generate much finer and individualized data to make the entire regularization more flexible. Furthermore, MSC is more general and can be applied to all kinds of networks through the proposed self-identity structure. The experimental results on all the benchmark datasets showed that the proposed MSC regularization method achieves the best performances in all the networks. [ABSTRACT FROM AUTHOR]

Pore-based fingerprint recognition has been researched for decades. Many algorithms have been proposed to improve the recognition accuracy of the system. However, the accuracies are always improved at the cost of speed. This article proposes a novel method to compare the pores in high-resolution fingerprint images using the popular coarse-to-fine strategy. A multiple spatial pairwise local co-occurrence descriptor is proposed to improve the calculation of the similarities between pores. It calculates multiple local co-occurrence statistics for each pore using its neighbors. The proposed method can establish correspondences between pores more accurately. The refinement of the correspondences is then achieved by using a local topology-preserving matching algorithm. The algorithm uses rotational invariant local structures and pore pair local topology similarities to calculate the cost of each correspondence. It can remove the mismatches more accurately and efficiently. The experimental results on two high-resolution fingerprint image databases show that the proposed algorithm perform well in both accuracy and speed comparing to the existing algorithms. [ABSTRACT FROM AUTHOR]

Defocus blur detection (DBD), which has been widely applied to various fields, aims to detect the out-of-focus or in-focus pixels from a single image. Despite the fact that the deep learning based methods applied to DBD have outperformed the hand-crafted feature based methods, the performance cannot still meet our requirement. In this paper, a novel network is established for DBD. Unlike existing methods which only learn the projection from the in-focus part to the ground-truth, both in-focus and out-of-focus pixels, which are completely and symmetrically complementary, are taken into account. Specifically, two symmetric branches are designed to jointly estimate the probability of focus and defocus pixels, respectively. Due to their complementary constraint, each layer in a branch is affected by an attention obtained from another branch, effectively learning the detailed information which may be ignored in one branch. The feature maps from these two branches are then passed through a unique fusion block to simultaneously get the two-channel output measured by a complementary loss. Additionally, instead of estimating only one binary map from a specific layer, each layer is encouraged to estimate the ground truth to guide the binary map estimation in its linked shallower layer followed by a top-to-bottom combination strategy, gradually exploiting the global and local information. Experimental results on released datasets demonstrate that our proposed method remarkably outperforms state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

Intensity inhomogeneity and noise are two common issues in images but inevitably lead to significant challenges for image segmentation and is particularly pronounced when the two issues simultaneously appear in one image. As a result, most existing level set models yield poor performance when applied to this images. To this end, this paper proposes a novel hybrid level set model, named adaptive variational level set model (AVLSM) by integrating an adaptive scale bias field correction term and a denoising term into one level set framework, which can simultaneously correct the severe inhomogeneous intensity and denoise in segmentation. Specifically, an adaptive scale bias field correction term is first defined to correct the severe inhomogeneous intensity by adaptively adjusting the scale according to the degree of intensity inhomogeneity while segmentation. More importantly, the proposed adaptive scale truncation function in the term is model-agnostic, which can be applied to most off-the-shelf models and improves their performance for image segmentation with severe intensity inhomogeneity. Then, a denoising energy term is constructed based on the variational model, which can remove not only common additive noise but also multiplicative noise often occurred in medical image during segmentation. Finally, by integrating the two proposed energy terms into a variational level set framework, the AVLSM is proposed. The experimental results on synthetic and real images demonstrate the superiority of AVLSM over most state-of-the-art level set models in terms of accuracy, robustness and running time. [ABSTRACT FROM AUTHOR]

In this article, two transmissive metasurfaces with different phase control mechanisms and working polarization condition are proposed to efficiently generate zero order Bessel beams (BBs), respectively. For the linear polarization condition, a dual-mode meta-atom constructed by three layers of dielectric substrate and four layers of metal patch is designed. By changing the size of metal patches of meta-atom along x-axis and y-axis, respectively, the phase shift for x- and y-polarized incident wave can be tuned relative independently of 0–2 π while the transmission coefficient keeps high value at 10 GHz. For the circular polarization condition, a sandwich type meta-atom applying geometric phase to realize full phase control while keep high efficiency near 10 GHz is designed. According to the designed phase profile of BBs, select meta-atoms with certain phase and efficiency, then arrange them to construct metasurfaces. Full wave simulation results indicate that the dual-mode metasurface working in linear polarization condition generates BBs with the maximum transmission efficiency 0.79 and 0.85 for x and y polarization wave, respectively, and the geometric phase metasurface working in circular polarization condition generates BBs with the maximum transmission efficiency 0.92. Besides, the propagation characteristics of the excited BBs are studied. [ABSTRACT FROM AUTHOR]

Multiview learning has been widely studied in various fields and achieved outstanding performances in comparison to many single-view-based approaches. In this paper, a novel multiview learning method based on the Gaussian process latent variable model (GPLVM) is proposed. In contrast to existing GPLVM methods which only assume that there are transformations from the latent variable to the multiple observed inputs, our proposed method simultaneously takes a back constraint into account, encoding multiple observations to the latent variable by enjoying the Gaussian process (GP) prior. Particularly, to overcome the difficulty of the covariance matrix calculation in the encoder, a linear projection is designed to map different observations to a consistent subspace first. The obtained variable in this subspace is then projected to the latent variable in the manifold space with the GP prior. Furthermore, different from most GPLVM methods which strongly assume that the covariance matrices follow a certain kernel function, for example, radial basis function (RBF), we introduce a multikernel strategy to design the covariance matrix, being more reasonable and adaptive for the data representation. In order to apply the presented approach to the classification, a discriminative prior is also embedded to the learned latent variables to encourage samples belonging to the same category to be close and those belonging to different categories to be far. Experimental results on three real-world databases substantiate the effectiveness and superiority of the proposed method compared with state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

In this contribution, a recently conducted measurement campaign for indoor millimeter-wave (mm-wave) propagation channels is introduced. A vector network analyzer (VNA)-based channel sounder was exploited to record the channel characteristics at the frequency band from 28–30 GHz. A virtual uniform circular array (UCA) with a radius of 0.25 m was formed using a rotator with 360 steps. Moreover, by taking advantage of fiber-optic technique applied in the channel sounder, measurements at 50 positions were performed from an indoor hall to an indoor corridor along a long pre-defined route. A low-complexity high-resolution propagation estimation (HRPE) algorithm is exploited to estimate the propagation parameters of multipath components (MPCs). Based on the HRPE estimation results, a novel clustering identification and tracking algorithm is proposed to trace clusters. Composite channel characteristics, cluster-level characteristics and dynamic (or birth-death) behaviours of the clusters are investigated, which constitute a dynamic model for the indoor mm-wave channel. [ABSTRACT FROM AUTHOR]

In this paper, a novel deep network is proposed for multi-focus image fusion, named Deep Regression Pair Learning (DRPL). In contrast to existing deep fusion methods which divide the input image into small patches and apply a classifier to judge whether the patch is in focus or not, DRPL directly converts the whole image into a binary mask without any patch operation, subsequently tackling the difficulty of the blur level estimation around the focused/defocused boundary. Simultaneously, a pair learning strategy, which takes a pair of complementary source images as inputs and generates two corresponding binary masks, is introduced into the model, greatly imposing the complementary constraint on each pair and making a large contribution to the performance improvement. Furthermore, as the edge or gradient does exist in the focus part while there is no similar property for the defocus part, we also embed a gradient loss to ensure the generated image to be all-in-focus. Then the structural similarity index (SSIM) is utilized to make a trade-off between the reference and fused images. Experimental results conducted on the synthetic and real-world datasets substantiate the effectiveness and superiority of DRPL compared with other state-of-the-art approaches. The source code can be found in https://github.com/sasky1/DPRL. [ABSTRACT FROM AUTHOR]

Traditional clinical experiences have shown the benefit of lesion location attention for improving clinical diagnosis tasks. Inspired by this point of interest, in this paper we propose a novel lesion location attention guided network named LLAGnet to focus on the discriminative features from lesion locations for multi-label thoracic disease classification in chest X-rays (CXRs). By revealing the equivalence of the region-level attention (RLA) and channel-level attention (CLA), we find that the RLA is available as priors for object localization while the CLA implicitly provides high weights to the attractive channels, which both enable lesion location attention excitation. To integrate the advantages from both mechanisms, the proposed LLAGnet is structured with two corresponding attention modules, i.e., the RLA and CLA modules. Specifically, the RLA module consists of the global and local branches. And the weakly supervised attention mechanism embedded in the global branch can obtain visual regions of lesion locations by back-propagating gradients. Then the optimal attention region is amplified and applied to the local branch to provide more fine-grained features for the image classification. Finally, the CLA module adaptively enhances the weights of channel-wise features from the lesion locations by modeling interdependencies among channels. Extensive experiments on the ChestX-ray14 dataset clearly substantiate the effectiveness of LLAGnet as compared with the state-of-the-art baselines. [ABSTRACT FROM AUTHOR]

Antenna correlation is an important measure for designing multiple-input–multiple-output (MIMO) antenna systems. A lower antenna correlation indicates a better MIMO performance that can be achieved with the underlying antenna systems. In the antenna design community, it is very common to evaluate the antenna correlation with isotropic or nonisotropic (e.g., Gaussian-distributed) angular power spectrum (APS) as baselines. On the other hand, antenna correlation can also be evaluated via channel transfer function (CTF) under the given propagation channel, e.g., drawn from the bidirectional geometry-based stochastic channel model. In this letter, the analytic forms for the antenna correlation based on the APS and the CTF are derived, respectively, with their similarities and differences explained. Moreover, a numerical example is also given with a standard channel model to support our findings. [ABSTRACT FROM AUTHOR]

Multiview learning methods often achieve improvement compared with single-view-based approaches in many applications. Due to the powerful nonlinear ability and probabilistic perspective of Gaussian process (GP), some GP-based multiview efforts were presented. However, most of these methods make a strong assumption on the kernel function (e.g., radial basis function), which limits the capacity of the real data modeling. In order to address this issue, in this paper, we propose a novel multiview approach by combining a multikernel and GP latent variable model. Instead of designing a deterministic kernel function, multiple kernel functions are established to automatically adapt various types of data. Considering a simple way of obtaining latent variables at the testing stage, a projection from the observed space to the latent space as a back constraint has also been simultaneously introduced into the proposed method. Additionally, different from some existing methods which apply the classifiers off-line, a hinge loss is embedded into the model to jointly learn the classification hyperplane, encouraging the latent variables belonging to the different classes to be separated. An efficient algorithm based on the gradient decent technique is constructed to optimize our method. Finally, we apply the proposed approach to three real-world datasets and the associated results demonstrate the effectiveness and superiority of our model compared with other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

A modified range-Doppler algorithm (M-RDA) for processing high squint synthetic aperture radar (SAR) echoes is proposed in this letter. Unlike the original RDA, the M-RDA directly corrects the range cell migration in the range frequency domain without interpolation, which greatly enhances its computational efficiency. When analyzing a high squint SAR echo signal, the approximation of the distance between the SAR antenna and the scene of interest leads to error in azimuthal compression. To address this problem, a fourth-order phase model is considered in the proposed algorithm to achieve accurate azimuthal compression. Finally, geometry correction is performed to obtain the final focused SAR images. The simulation results show that the proposed algorithm can process high squint SAR echoes effectively. [ABSTRACT FROM AUTHOR]

The theoretical performance of massive multiple-input multiple-output (MIMO) depends on the favorable propagation condition (FPC), where user channels are asymptotically orthogonal. Recent researches have demonstrated that the FPC could be approached in real propagation scenarios, while the gap between a real propagation channel and the ideal channel has not been further modeled. This letter proposes an empirical formula for the ratio of sum rate capacity to its upper bound based on field measurement in indoor and outdoor scenarios. The formula has a simple expression with the number of antennas in the array, the number of users and a scenario-dependent parameter. The proposed formula could be employed to evaluate the “favorable” extent for the massive MIMO channel. [ABSTRACT FROM PUBLISHER]

A complete facet model for the backscattering from rough sea surface with breaking waves is proposed in this paper. In consideration of the spatial distribution of breaking waves on sea surface, the model is able to give a good interpretation to the “super events” under high sea states, such as sea spikes and high polarization ratios. Based on the proposed model, normalized radar cross section plots of sea surface under backscattering configuration are calculated and compared with measured data. The comparisons show that the proposed model is tractable to estimate the scattering from electrically large ocean surface under high sea states with accuracy and efficiency. In addition, the non-Gaussian statistics and spatial correlation properties of sea clutter are analyzed under different range resolutions and incident angles. The results show that high kurtosis value, which means a sea spike phenomenon, mostly happens in lower grazing angle and higher range resolution cases. The comparisons of simulated SAR images of sea surface with and without breaking waves also reveal that sea spikes and high polarization ratios are caused by breaking waves. All the simulation results prove that our model not only is able to explain the physical mechanism of the scattering but also can be applied to the analyses of statistical properties of sea clutter under high sea states. [ABSTRACT FROM PUBLISHER]

An improved facet model for the predication of the normalized radar cross section (NRCS) of electrically large rough sea surface was proposed based on the first-order small slope approximation (SSA-1) method, the Bragg scattering mechanism, and the specular scattering mechanism. The proposed method is able to evaluate both the complex reflective function and NRCS of electrically large sea surfaces from as low as ultrahigh frequency band to as high as Ka-band. The main idea is that a tilt sea facet can be regarded as the superposition of a planner facet and the microscopic profile; the latter is assumed to be a set of sinusoidal ripple patches. Thus, the integration kernel in SSA-1 over several small facets can be replaced by a large facet with a short wave modification. The efficiency increases because of the much larger mesh size than SSA. Then, both the backscattering and bistatic scattering NRCS results calculated by the proposed method were compared with those predicated by SSA-1, and all of the results show that the proposed method has the merits of high calculation efficiency as well as calculation accuracy. [ABSTRACT FROM PUBLISHER]