In this work, we introduce a dual transformation network for single image contrast enhancement, which usually aims to improve global contrast and enrich local details. To this end, we propose two parallel branches to respectively handle the two goals by learning different kinds of transformations. Specifically, one branch aims to construct a global transformation curve to improve global contrast, while

Recent end-to-end (e2e) speech synthesis systems usually employ attention techniques to align an input text sequence against a mel-spectrogram sequence. Attention-based e2e approach has shown state-of-the-art performance in speech synthesis. However, generating stable and robust attention alignment to avoid some serious failures such as repeating, missing, and mumbling phones is still an ongoing challenge

This letter proposes a multi-frame super-resolution framework by combining Demons registration with Bayesian-based regularized reconstruction. For both of our proposals, D-BTVIR and D-IRWIR, the visual analysis shows improvements in regions where the compared methods produced results with either motion-artifacts or over-smoothed aspect due to misregistration. Quantitative results on simulated deformations

As a smart visual attention mechanism to analyze visual scenes, visual saliency has been shown to closely correlate with semantic information such as faces. Although many semantic-information-guided saliency models have been proposed, to the best of our knowledge, no semantic information in affective domain has been employed for saliency detection. Aesthetic, the affective perceptual quality that integrates

Recurrent neural network transducers (RNN-T) are a promising end-to-end speech recognition framework that transduce input acoustic frames to a character sequence. Best- and breadth-first searches have been used as decoding strategies for RNN-T. However, best-first search follows a sequential process for its expansion search, which slows down the decoding process. Although breadth-first search replaces

Convolutional neural networks (CNNs), which have been the state-of-the-art in many image-related applications, are prone to losing important spatial information between image instances. Capsule networks (CapsNets), on the other hand, are capable of leveraging such information through their routing by agreement process, making them powerful architectures for small datasets, such as medical imaging ones

The diagnosis of myocardial infarction (MI) in the presence of other cardiac diseases having similar electrocardiogram (ECG) characteristics that of MI is a challenging problem. Existing automated methods have used the standard 12-lead ECG to detect MI from healthy controls (HC). However, these methods may not provide reliable MI diagnosis in the presence of mimicking MI diseases. Therefore, in this

High-dimensional small sample size problems exist in the real world, which significantly increases the difficulty of data processing. In this paper, we propose a kernel-based within class collaborative preserving discriminant projection method to reduce data dimensionality. In order to deal with nonlinear problems and improve the discrimination of the projection subspace, the proposed method preserves

We propose a novel direction-of-arrival (DOA) estimation method for hybrid millimeter-wave massive MIMO systems at low SNR. Unlike the existing hierarchical-search-based methods that directly estimate the DOA from the angular power, we investigate the angular power gradient based on beamforming with incremental gains. Then, we leverage the difference of beamforming gain among codebook levels and adopt

There are many intelligent devices around us that have the ability to emit and receive audio signals, such as smartphones, smartwatches, and tablets. If they constitute an acoustic transceiver network, it can boost the performance of many audio processing tasks. The speaker localization and tracking algorithms using such a network require the prior information of node positions. To acquire this knowledge

In this letter, we devise an adaptive double-layered embedding scheme that is suitable for MP3 steganography. According to the encoding characteristics of MP3, the linbits are used as the embedding domain. The proposed scheme is divided into two layers and the messages are embedded into each layer with binary STCs. The cost function used in the first layer is designed by employing the masking effect

We develop Adaptive Cosine Estimator (ACE) type detector for non-zero mean Gaussian interference specifically for the replacement and additive target models of the hyperspectral imaging problem. We consider the case where the data under test and the training samples differ from one scaling factor on the mean and one scaling factor on the covariance matrix. We derive two-step generalized likelihood

Multi-label hashing is a new research topic in image retrieval. As images are usually associated with multiple semantic labels, there is multi-level semantic similarity such as very similar, normally similar and dissimilar among multi-label images. In order to obtain the multi-level semantic similarity, this letter constructs a hypergraph in label space by creating a hyperedge for each semantic label

Automatic facial expression recognition (FER) is a fundamental topic in computer vision. Many studies have indicated that facial emotion changes are strongly related to certain regions of interest (ROIs), such as the mouth, eyes, eyebrows, and nose; therefore, the features of these facial ROIs are very important for identifying expressions. Since Gabor filters are very efficient in extracting visual

This letter deals with the problem of robust beamforming in the presence of non-Gaussian impulsive noise. Under this framework, a new robust extension of the empirical MVDR beamformer is developed. The proposed extension is a plug-in estimate of a measure-transformed MVDR (MT-MVDR) beamformer, that operates by applying a transform to the probability distribution of the data. The considered transform

Shadow detection is one of the most challenging issues in computer vision. Inspired by the great success of the convolutional neural network (CNN) for the problem of image restoration, learned features have been widely adopted for shadow detection. However, most existing methods still suffer from ambiguities driven by black-colored objects, which are not actually shaded, as well as the background clutter

In order to maximize the signal-to-interference-plus-noise ratio (SINR) under a constant-envelope (CE) constraint, a fast and efficient joint design of the transmit waveform and the receive filter for colocated multiple-input multiple-output (MIMO) radars is essential. Conventional joint optimization is performed using nonlinear optimization techniques such as the semidefinite relaxation (SDR) algorithm

One of the key issues in photometric stereo is the extension of its application in real world objects which shows non-Lambertian reflectance. This letter proposes a multi-scale weighted convolutional fusion network with deep learning architecture to realize high-precision perception of non-Lambertian surfaces under arbitrary illumination conditions. A multi-scale convolutional fusion module is designed

In systems that aim to mitigate the inter-carrier interference (ICI) caused by doubly selective (DS) channels, the partial Fast Fourier Transform (PFFT) has emerged as an interesting alternative to the conventional FFT. In this letter, we propose a novel PFFT demodulation scheme based on fast convolution (FC) structure. The proposed scheme uses the overlap-save operation of FC to further suppress the

Monaural source separation is often conducted by manipulating the amplitude spectrogram of a mixture (e.g., via time-frequency masking and spectral subtraction). The obtained amplitudes are converted back to the time domain by using the phase of the mixture or by applying phase reconstruction. Although phase reconstruction performs well for the true amplitudes, its performance is degraded when the

In the field of 3D face alignment, most researchers have focused on improving the prediction accuracy of algorithms and ignored the portability for practical applications. To this end, this study presents a real-time 3D face-alignment method that uses an encoder-decoder network with an efficient deconvolution layer. The fusion of the encoding and decoding feature adds more abundant features to this

This work addresses a challenging problem of estimating the full 3D human shape and pose from monocular videos. Since real-world 3D mesh-labeled datasets are limited, most current methods in 3D human shape reconstruction only focus on single RGB images, losing all the temporal information. In contrast, we propose temporally refined Graph U-Nets, including an image-level module and a video-level module

An efficient smoothing algorithm is presented in this letter. In this algorithm, while the signal is represented by sum of Gaussian functions-based dynamical model, its states are estimated using a constrained optimization problem where the Gaussian recurrence relation enters as constrain. The accuracy of the proposed method depends on the proper choice of regularization parameters. The value of the

This letter studies the legitimate surveillance with one half-duplex legitimate monitor (E) over the suspicious multichannel relaying system, where the suspicious transmitter (ST) and relay (SR) implement the optimal power allocation over all orthogonal channels under a joint sum-power constraint to maximize the communication rate of the suspicious system. Under this setup and considering that SR operates

This work revisits the problem of distributed adaptive filtering in multi-agent sensor networks. In contrast to classical approaches, the formulation relaxes the Gaussian assumption on the signal and noise to the generalized setting of $\alpha$ -stable distributions that do not possess second- and higher-order statistical moments. Most importantly, the considered scenario allows for different characteristic

The Listen, Attend and Spell (LAS) model and other attention-based automatic speech recognition (ASR) models have known limitations when operated in a fully online mode. In this letter, we analyze the online operation of LAS models to demonstrate that these limitations stem from the handling of silence regions and the reliability of online attention mechanism at the edge of input buffers. We propose

As an effective tool for two-dimensional data analysis, two-dimensional canonical correlation analysis (2DCCA) is not only capable of preserving the intrinsic structural information of original two-dimensional (2D) data, but also reduces the computational complexity effectively. However, due to the unsupervised nature, 2DCCA is incapable of extracting sufficient discriminatory representations, resulting

Capsule projection network (CapProNet) has shown its ability to obtain semantic information, and spatial structural information from the raw images. However, the vector capsule of CapProNet has limitations in representing semantic information due to ignoring local information. Besides, the number of trainable parameters also increases greatly with the dimension of the feature vector. To that end, we

Binary sequences are widely used in various practical fields, such as telecommunications, radar technology, navigation, cryptography, measurement sciences, biology or industry. In this letter, a method to generate long binary sequences (LBS) with low peak sidelobe level (PSL) value is proposed. Having an LBS with length $n$ , both the time and memory complexities of the proposed algorithm are $\mathcal

This letter considers an $M$ -ary hypothesis testing problem on an $n$ -dimensional random vector perturbed by the addition of Gaussian noise. A novel expression for the gradient of the error probability, with respect to the covariance matrix of the noise, is derived and shown to be a function of the cross-covariance matrix between the noise matrix (i.e., the matrix obtained by multiplying the noise

In recent years, a large number of quantization schemes have been proposed for compressing convolutional neural networks (CNN). However, most of them have the following problems: 1) when there are outliers in the weight, post-training quantization cannot obtain the ideal effect, and the accuracy loss is unavoidable; 2) quantizing the non-maximum suppression (NMS) stage of oriented object detection

A novel approach for the parameter estimation of polynomial phase signals (PPSs) is proposed. Unlike the existing methods, the proposed method uses the spectrum phase (SP) rather than the traditional instantaneous phase (IP) for parameter estimation. First, we present an unwrapping procedure for the SP. Second, we derive a method that can convert the unwrapped SP into the IP and then perform polynomial

This leter develops a universal technique for analyzing discrete super-resolution algorithms with $\ell _1$ -norm based objective function. Though the super-resolution problem with sparsity constraints is of intense research interest in the past decade, only a modified Dantzig selector has been non-asymptotically analyzed without additional structural information whereas this theoretical guarantee

The identification of transversal filters is important for numerous applications. In acoustical applications a first order Markov model is often used to describe the time-variant nature of transversal filters. The Kalman filter is the optimal unbiased estimator for such a Markov model. It inherently calculates the uncertainty of the current state estimate by its state error covariance matrix. In contrast

Most existing speech separation research focuses on improving the separation performance under consistent source number conditions between training and testing. In real-world applications, however, the source number may be different from that in training sets. In this letter, we address this problem by thoroughly improving the deep attractor network in terms of the network architecture and learning

In weakly supervised object detection, most existing approaches are proposed for images. Without box-level annotations, these methods cannot accurately locate objects. Considering an object may show different motion from its surrounding objects or background, we leverage motion information to improve the detection accuracy. However, the motion pattern of an object is complex. Different parts of an

In this letter, we present a novel rotation averaging pipeline, which is performed in a hierarchical manner. Unlike the traditional rotation averaging methods which focus on designing robust loss function to get rid of the impacts of the relative rotation outliers, here the outliers are detected and filtered by leveraging the well-known robust model estimation procedure, RANdom SAmple Consensus (RANSAC)

In this letter, we propose a vein-based biometric verification system relying on deep learning. A novel approach consisting of a convolutional neural network (CNN), trained in a supervised manner, cascaded with an auto-encoder, trained in an unsupervised way, is here exploited. In more detail, a novel densely-connected convolutional autoencoder is here used on top of backbone CNNs. This architecture

This paper proposes a covariance matrix completion based algorithm for underdetermined direction of arrival (DOA) estimation in the presence of unknown nonuniform noise using nested sparse circular array (NSCA) with only $N$ sensors. The proposed algorithm provides a systematic procedure to complete a covariance matrix for a virtual uniform circular array (UCA) with $M$ sensors ( $M > N$ ). Compared

Graph filters and their inverses have been widely used in denoising, smoothing, sampling, interpolating and learning. Implementation of an inverse filtering procedure on spatially distributed networks (SDNs) is a remarkable challenge, as each agent on an SDN is equipped with a data processing subsystem with limited capacity and a communication subsystem with confined range due to engineering limitations

We propose a privacy-preserving distributed maximum consensus algorithm where the local state of the agents and identity of the maximum state owner is kept private from adversaries. To that end, we reformulate the maximum consensus problem over a distributed network as a linear program. This optimization problem is solved in a distributed manner using the alternating direction method of multipliers

Deep learning-based methods have achieved remarkable performance in image enhancement but generally require huge GPU memory, and computational cost when enhancing the high-resolution images. We explore to enhance high-resolution images like manual retouching by digital artists to obtain stable, and excellent performance in real-time. We extract the features from their thumbnails, and utilizes these

Tensor completion (TC) attempts to estimate the missing tensor entries from partial observations. Over the past decade, a tremendous amount of work has focused on this problem, which has given rise to a few excellent TC methods. Thus, a critical question is how to verify the optimality of TC models, especially when the underlying tensor is unknown, which is a common scenario in practice. The only existing

In this work, we develop and compare two innovative strategies for parameter estimation and radar detection of multiple point-like targets. The first strategy, which appears here for the first time, jointly exploits the maximum likelihood approach and Bayesian learning to estimate targets’ parameters including their positions in terms of range bins. The second strategy relies on the intuition that

This letter proposes a double-attentive principle component analysis (DA-PCA) model for image processing. Compared to the previous PCA-based works that cannot deal with normal images and outliers effectively, the proposed DA-PCA model performs a double-attentive mechanism to sever the connections with outliers and hold the effectiveness of normal images. To solve the proposed DA-PCA model, we propose

Salient object detection (SOD) in 360 $^{\circ }$ omnidirectional images has become an eye-catching problem because of the popularity of affordable 360 $^{\circ }$ cameras. In this paper, we propose a Features Adaptation Network (FANet) to highlight salient objects in 360 $^{\circ }$ omnidirectional images reliably. To utilize the feature extraction capability of convolutional neural networks and capture

The auxiliary normal equation is proposed to construct an incremental update (IU) system in which the increment of weight vector rather than the weight itself is optimized at each iteration, which however, can only deal with problems under linearity assumption. In this letter, we establish the IU system in a fixed-dimension feature space induced by the random Fourier features mapping, enabling it to

This paper considers symbol-level precoding (SLP) for multiuser multi-input single-output (MISO) downlink transmission with the aid of intelligent reflecting surface (IRS). Specifically, by assuming one-bit transmitted signals at the base station (BS), which arises from the use of low-resolution DACs in the regime of massive transmit antennas, a joint design of one-bit SLP at the BS and the phase shifts

Most anti-collusion audio fingerprinting schemes are aiming at finding colluders from the illegal redistributed audio copies. However, the loss caused by the redistributed versions is inevitable. In this letter, a novel fingerprinting scheme is proposed to eliminate the motivation of collusion attack. The audio signal is transformed to the frequency domain by the Fourier transform, and the coefficients

High frame rate videos are increasingly getting popular in recent years, driven by the strong requirements of the entertainment and streaming industries to provide high quality of experiences to consumers. To achieve the best trade-offs between the bandwidth requirements and video quality in terms of frame rate adaptation, it is imperative to understand the effects of frame rate on video quality. In

In frequency division duplex (FDD) multiple-input multiple-output (MIMO) wireless communications, limited channel state information (CSI) feedback is a central tool to support advanced single- and multi-user MIMO beamforming/precoding. To achieve a given CSI quality, the CSI quantization codebook size has to grow exponentially with the number of antennas, leading to quantization complexity, as well

This letter focuses on image manipulation detection which aims to recognize the manipulated regions under the contextual semantic information. Existing approaches usually overlook the semantic discrepancy between different levels of feature maps, and directly fuse (e.g., addition, or concatenation) them for detection. In this letter, we argue that the semantic gap is the main reason for the low effectiveness

This letter develops a novel Continuous-Time Accelerated Proximal Point Algorithm (CAPPA) for $\ell _1$ -minimization problems with provable fixed-time convergence guarantees. The problem of $\ell _1$ -minimization appears in several contexts such as Sparse Recovery (SR) in Compressed Sensing (CS) theory and sparse linear and logistic regressions in machine learning. Most existing algorithms for solving

This letter is focused on the problem of estimating the two parameters of a uniform distribution — its support boundaries — and the application of the Cramér-Rao-Leibniz Lower Bound, a replacement for the Cramér-Rao Lower Bound when the latter does not hold.

A proper representation that can well express the characteristics of a word plays an important role in wake-up word detection (WWD). However, it may be easily corrupted due to various types of environmental noise occurred in the place where WWD typically works, causing unreliable performance. To deal with this practical issue, we propose a novel strategy called cross-informed domain adversarial training

In this letter, for outage performance orientated large intelligent surfaces (LISs)-assisted point to point wireless systems with severely blocked direct link and Rayleigh fading channels,we first propose a jointly interleaved training and transmission design. Then a semi-closed form expression is derived for the average training overhead. And it is shown to be upper bounded by the minimum between

A wide variety of activation functions have been proposed for neural networks. The Rectified Linear Unit (ReLU) is especially popular today. There are many practical reasons that motivate the use of the ReLU. This paper provides new theoretical characterizations that support the use of the ReLU, its variants such as the leaky ReLU, as well as other activation functions in the case of univariate, single-hidden

Multitarget tracking (MTT) is a challenging task that aims at estimating the number of targets and their states from measurements provided by one or multiple sensors. Additional information, such as imperfect estimates of target classes provided by a classifier, can facilitate the target-measurement association and thus improve MTT performance. In this letter, we describe how a recently proposed MTT

Gaussian belief propagation (BP) is widely used for distributed inference. Its computational complexity, and communication overhead depend on the total number of messages updated, and transmitted among variable nodes, respectively. For large, and dense networks, both computational complexity, and communication overhead could be extremely high. To this end, a variant of Gaussian BP called Gaussian SPAWN

High-accuracy real-time Doppler estimation is one of the key problems in underwater acoustic communication systems. Using the dual-hyperbolic-frequency-modulation (dual-HFM) synchronous signal is an effective solution; however, the Doppler estimation accuracy is limited by the sampling frequency in the traditional methods. In this letter, we propose a high-accuracy speed spectrum scanning method based