In this letter, we present an attentional non-linear activation function approximation method called ATA for VLSI-based neural networks. Unlike other approximation methods that pursue the low hardware resources with a high recognition accuracy loss, the ATA utilizes the pixel attention to focus on the important features to keep the recognition accuracy and reduce resource cost. Specifically, attention

In this letter, a novel angular jerk estimator (AJE) in the time-frequency (TF) domain is proposed. For this purpose, the properties of the short-time Fourier transform (STFT) are taken into consideration, which allows for computation of the instantaneous angular jerk for each time and frequency $\boldsymbol{(t,\omega)}$ coordinate. Additionally, apart from the introduced AJE, two novel forms of the

End-to-end multi-talker speech recognition is an emerging research trend in the speech community due to its vast potential in applications such as conversation and meeting transcriptions. To the best of our knowledge, all existing research works are constrained in the offline scenario. In this work, we propose the Streaming Unmixing and Recognition Transducer (SURT) for end-to-end multi-talker speech

A fog-radio access network (F-RAN) architecture is studied for an Internet-of-Things (IoT) system in which wireless sensors monitor a number of multi-valued events and transmit in the uplink using grant-free random access to multiple edge nodes (ENs). Each EN is connected to a central processor (CP) via a finite-capacity fronthaul link. In contrast to conventional information-agnostic protocols based

Modeling various aspects that make a music piece unique is a challenging task, requiring the combination of multiple sources of information. Deep learning is commonly used to obtain representations using various sources of information, such as the audio, interactions between users and songs, or associated genre metadata. Recently, contrastive learning has led to representations that generalize better

Representing fingerprint templates in binary form can provide outstanding merits compared to the conventional minutiae-based fingerprint recognition system. The existing fixed-length fingerprint feature extraction methods either suffer from redundant feature magnitude or lack of template security. In this letter, we present a compact (128 bytes) and cancelable fingerprint binary codes generation scheme

The research of part-based methods has been proven as an effective way in person re-identification (Re-ID) task. However, in existing part-based Re-ID methods, the informative interactions and potential associations among parts are neglected, which demands further study. To fill this gap, we propose a novel Part-relation-aware Feature Fusion Network (PFFN) which achieves a part-level feature fusion

Solving complex optimization problems in engineering and the physical sciences requires repetitive computation of multi-dimensional function derivatives, which commonly require computationally-demanding numerical differentiation such as perturbation techniques. In particular, Gauss-Newton methods are used for nonlinear inverse problems that require iterative updates to be computed from the Jacobian

Automatic analysis of biomedical signals plays an important role in the auxiliary diagnosis of diseases. Traditional methods extract hand-crafted features by imitating doctors’ experience, while recent methods focus on extracting deep features automatically by designing the architectures of deep neural networks (DNNs). Combining these two kinds of features can not only take advantage of doctors’ experience

The fully convolutional network (FCN) has achieved tremendous success in dense visual recognition tasks, such as scene segmentation. The last layer of FCN is typically a global classifier ( $1\times 1$ convolution) to recognize each pixel to a semantic label. We empirically show that this global classifier, ignoring the intra-class distinction, may lead to sub-optimal results. In this work, we present

In this letter, we propose an end-to-end stitching network, which takes two images with a narrow field of view (FOV) as inputs, and produces a single image with a wide FOV. Our method estimates multiple homographies to cover the depth differences in the scene and is therefore robust against parallax distortion. In particular, global warping maps are generated using estimated multiple homographies and

Near-infrared to visible (NIR-VIS) face recognition is the most common case in heterogeneous face recognition, which aims to match a pair of face images captured from two different modalities. Existing deep learning based methods have made remarkable progress in NIR-VIS face recognition, while it encounters certain newly-emerged difficulties during the pandemic of COVID-19, since people are supposed

Deception against inverse synthetic aperture radar (ISAR) has been a topic of active research in electronic warfare. However, most of the studies mainly focus on producing one-dimensional deception, which is inadequate for two-dimensional imaging system like ISAR. Thus, in this letter, a joint two-dimensional ISAR deception based on frequency diverse array (FDA) and interrupted sampling is proposed

This paper proposes a decision scheme based on the $k$ -nearest neighbors rule to detect coherent signals in non-Gaussian noise modeled as the sum of K-distributed clutter plus thermal noise. The analysis is conducted also on real data recordings and shows that the proposed detector can outperform natural competitors.

In this paper, we discuss a new approach to direction of arrival estimation for systems with subarray sampling. We propose to estimate the covariance matrix of the full array from the sample covariance matrices of the subarrays using a neural network. This technique enables the estimation of more sources than radio frequency chains by applying a MUSIC estimator to the reconstructed full covariance

End-to-end models have achieved impressive results on the task of automatic speech recognition (ASR). For low-resource ASR tasks, however, labeled data can hardly satisfy the demand of end-to-end models. Self-supervised acoustic pre-training has already shown its impressive ASR performance, while the transcription is still inadequate for language modeling in end-to-end models. In this work, we fuse

We propose a novel lazy-evaluation token-group decoding algorithm with on-the-fly composition of weighted finite-state transducers (WFSTs) for large vocabulary continuous speech recognition. In the standard on-the-fly composition decoder, a base WFST and one or more incremental WFSTs are composed during decoding, and then token passing algorithm is employed to generate the lattice on the composed search

Due to the significant variability in waveforms and characteristics of ECG signals, developing fully automatic (i.e., requires no expert assistance) heartbeat classification algorithms with satisfactory performance on domain-shifted data remains challenging. In this letter, we propose a novel Mixup Asymmetric Tri-training (MIAT) method to improve the generalization ability of heartbeat classifiers

In this letter, we consider the policy evaluation problem with linear function approximation in the context of decentralized multi-agent reinforcement learning (MARL), where the agents with a fixed joint policy cooperate to estimate the global expected accumulative reward through a decentralized communication network. In the existing algorithms, every agent updates its local parameter by combining

This letter addresses the detection problem of dim maneuvering targets in the presence of range cell migration. Specifically, it is assumed that the moving target can appear in more than one range cell within the transmitted pulse train. Then, the Bayesian information criterion and the generalized likelihood ratio test design procedure are jointly exploited to come up with six adaptive decision schemes

Loss function plays a key role in self-supervised monocular depth estimation methods. Current reprojection loss functions are hand-designed and mainly focus on local patch similarity but overlook the global distribution differences between a synthetic image and a target image. In this paper, we leverage global distribution differences by introducing an adversarial loss into the training stage of self-supervised

Fine-grained image-text retrieval aims at searching relevant images among fine-grained classes given a text query or in a reverse way. The challenges are not only bridging the gap between two heterogeneous modalities but also dealing with large inter-class similarity and intra-class variance existed in fine-grained data. To deal with the above challenges, we propose a Discriminative Latent Space Learning

The reconstruction of sparse signal is an active area of research. Different from a typical i.i.d. assumption, this paper considers a non-independent prior of group structure. For this more practical setup, we propose EM-aided HyGEC, a new algorithm to address the stability issue and the hyper-parameter issue facing the other algorithms. The instability problem results from the ill condition of the

This letter presents a conceptually simple and effective dual-granularity triplet loss for visible-thermal person re-identification (VT-ReID). Generally, ReID models are always trained with the sample-based triplet loss and identification loss from the fine granularity level. Further, center-based loss could be introduced to encourage the intra-class compactness and inter-class discrimination from

Exploiting both RGB (2D appearance) and Depth (3D geometry) information can improve the performance of semantic segmentation. However, due to the inherent difference between the RGB and Depth information, it remains a challenging problem in how to integrate RGB-D features effectively. In this letter, to address this issue, we propose a Non-local Aggregation Network (NANet), with a well-designed Multi-modality

The high redundancy among keyframes is a critical issue for the prior summarizing methods in dealing with user-created videos. To address the critical issue, we present a Graph Attention Networks (GAT) adjusted Bi-directional Long Short-term Memory (Bi-LSTM) model for unsupervised video summarization. First, the GAT is adopted to transform an image's visual features into higher-level features by the

Although great process has been achieved to accurately classify images, many methods only use labeled images while ignoring the large quantity of unlabeled images. To make use of unlabeled images, in this letter, we propose a novel transductive visual recognition method using the prediction consistency of multiple views (T-PCMV). Both labeled and unlabeled images are used in a unified framework. The

Target masking due to high sidelobes in the range-Doppler domain is a common problem in radar signal processing. Masking occurs when there is large variation in targets’ signal-to-noise-ratio levels, caused by differences in either range or radar cross section. In this letter, we consider the design of a waveform to reduce the masking of far range targets caused by near range targets. Our method is

Depth estimation from a single image is a crucial but challenging task for reconstructing 3D structures and inferring scene geometry. However, most existing methods fail to extract more detailed information and estimate the distant small-scale objects well. In this paper, we propose a monocular depth estimation based on multi-scale feature fusion. Specifically, to obtain input features of different

This letter develops an easily-implementable version of Page's CUSUM quickest-detection test, designed to work in certain composite hypothesis scenarios with time-varying data statistics. The decision statistic can be cast in a recursive form and is particularly suited for on-line analysis. By back-testing our approach on publicly-available COVID-19 data we find reliable early warning of infection

Unknown array non-ideal characteristics, including mutual coupling between array elements, channel gain/phase errors, and sensor position errors, will degrade the array performance of direction finding and beamforming. In this letter, we propose an improved maximum likelihood (ML) calibration algorithm to estimate all such non-ideal characteristics with lower computational complexity and faster convergence

Both the Hilbert transform and the B-spline wavelets are important tools in signal processing, which makes a study of relation between these two subjects be of practical significance. In particular, because the B-spline wavelets have good properties including vanishing moments, symmetry, compact support and so on, we focus on the Hilbert transform of B-spline wavelets in this letter. For this purpose

Continuous dimensional emotion recognition for facial video sequence is a crucial and challenging task in Affective Computing and Human-Computer Intelligent Interaction. The key of this task is to effectively extract and discriminate spatial-temporal features in a more fine-grained way. In this paper, a Two-Stage Spatiotemporal Attention Temporal Convolution Network (TS-SATCN) is designed for continuous

Self-supervised pre-training using so-called “pretext” tasks has recently shown impressive performance across a wide range of modalities. In this work, we advance self-supervised learning from permutations, by pre-training a model to reorder shuffled parts of the spectrogram of an audio signal, to improve downstream classification performance. We make two main contributions. First, we overcome the

Importance sampling (IS) is a powerful Monte Carlo (MC) methodology for approximating integrals, for instance in the context of Bayesian inference. In IS, the samples are simulated from the so-called proposal distribution, and the choice of this proposal is key for achieving a high performance. In adaptive IS (AIS) methods, a set of proposals is iteratively improved. AIS is a relevant and timely methodology

In this letter, we formulate the adaptive beamforming problem in terms of the constrained minimization of the Gaussian entropy of the array output signal, subject to the distortionless constraint of the signal of interest (SOI). We develop a quasi-steepest-descent beamformer as well as a stochastic gradient beamformer for the reception of a second-order non-circular/circular SOI. We also derive the

Word embeddings such as ELMo and BERT have been shown to model word usage in language with greater efficacy through contextualized learning on large-scale language corpora, resulting in significant performance improvement across many natural language processing tasks. In this work we integrate acoustic information into contextualized lexical embeddings through the addition of a parallel stream to the

Bilateral filter has demonstrated its effectiveness in many traditional methods for image restoration tasks. In this letter, we incorporate the idea of bilateral grid processing in a CNN framework and propose a bilateral stereo super-resolution network (BSSRnet). Specifically, we use a parallax-attention module to incorporate information from left and right views to learn content-aware bilateral filters

In this letter, a novel weighted mean square error (WMSE) is proposed to improve the DNN-based mask approximation method for speech enhancement, in which the weighting is closely related to the power exponent about noisy spectrum amplitude (NSA) base. The power exponents 0 and 2 separately reflect ideal amplitude masking (IAM) without any clippings and the indirect mapping (IM) on short-time spectral

Nonparametric maximum likelihood estimation is intended to infer the unknown density distribution while making as few assumptions as possible. To alleviate the over parameterization in nonparametric data fitting, smoothing assumptions are usually merged into the estimation. In this letter a novel boosting-based method is introduced to the nonparametric estimation in univariate cases. We deduce the

The focus of this letter is the estimation of a delay between two signals. Such a problem is common in signal processing and particularly challenging when the delay is non-stationary in nature. Our proposed solution is based on an all-pass filter framework comprising of two elements: a time delay is equivalent to all-pass filtering and an all-pass filter can be represented in terms of a ratio of a

As the digital version of a continuous-time delay, the concept of fractional delay (FD) is exploited to approximate a desired delay that is not a multiple of the sampling interval. However, in FD filters, there is always a severe distortion at the beginning of delayed signals, referred to as head distortion . This letter identifies the cause of head distortion and proposes a solution to this problem

With the advent of efficient low-cost processors and electroacoustic components, there is renewed interest in the practical implementation of active noise control (ANC). However, the slow convergence of conventional adaptive algorithms deployed in ANC restricts its handling of typical amplitude-varying noise. Hence, we proposed a modified model-agnostic, meta-learning (MAML) strategy to obtain an initial

A common approach for synchronizing the agents in a Wireless Sensor Network (WSN) is two-way timing stamps exchange mechanism. In this letter, by analyzing the change of accumulated clock offset within each communication cycle, an improved two-state discrete-time state model is proposed. A step further, two Kalman-filter-based estimators are proposed, by which the clock parameters and propagation delays

In two-way time-of-arrival (TOA) systems, a user device (UD) obtains its position by round-trip communications to a number of anchor nodes (ANs) at known locations. The objective function of the maximum likelihood (ML) method for two-way TOA localization is nonconvex. Thus, the widely-adopted Gauss-Newton iterative method to solve the ML estimator usually suffers from the local minima problem. In this

First-order differential microphone arrays (FODMAs), which combine a small-spacing uniform linear array and a first-order differential beamformer, have been used in a wide range of applications for sound and speech signal acquisition. However, traditional FODMAs are not steerable and their main lobe can only be at the endfire directions. To circumvent this problem, we propose in this letter a new method

Epilepsy is the second most common neurological disorder, affecting 65 million people around the world. It is associated with seizures - a sudden, uncontrolled electrical disturbance in the brain that can lead to profound transient changes in behavior, movements, feelings, and levels of consciousness. Current approaches to developing a generalized automated seizure detection algorithm rely on constructing

Few-shot image classification with graph neural network (GNN) is a hot topic in recent years. Most GNN-based approaches have achieved promising performance. These methods utilize node features or one-dimensional edge feature for classification ignoring rich edge featues between nodes. In this letter, we propose a novel graph neural network exploiting multi-dimensional edge features (MDE-GNN) based

A style-based generative adversarial network (StyleGAN2) yields remarkable results in image-to-latent embedding. This work proposes a Detached Dual-channel Domain Encoder as an effective and robust method to embed an image to a latent code, i.e., GAN inversion. It infers a latent code from two aspects: a) a detached dual-channel design to support faithful image reconstruction; and b) a local skip connection

Recent linguistic steganalysis methods model texts as sequences and use deep learning models to extract discriminative features for detecting the presence of secret information in texts. However, natural language has a complex syntactic structure and sequences have limited representation ability for text modeling. Moreover, previous methods tend to extract features from local continuous word sequences

Deep hashing has greatly improved retrieval performance with the powerful learning capability of deep neural network. However, deep unsupervised hashing can hardly achieve impressive performance due to the lack of the semantic supervision. This letter proposes Unsupervised Discriminative Deep Hashing (UD $^2$ H) to fulfill this gap. UD $^2$ H is formulated to jointly perform hash code learning and

Traditional adversarial attacks rely upon the perturbations generated by gradients from the network which are generally safeguarded by gradient guided search to provide an adversarial counterpart to the network. In this letter, we propose a novel framework to generate adversarial examples where the actual image is not corrupted rather its latent space representation is utilized to tamper the inherent

While integrated sensing and communication has attracted great attention, integrated imaging and communication is still largely open. Microwave, millimeter-wave, and tera-Hz imaging needs to sample a huge number of antenna positions and is thus too complex and costly to mobile communication devices. To address this challenge, we propose a Lightweight Imaging Algorithm (LIA) to reduce the complexity

To effectively fuse speech and visual features, this letter proposes a multi-modal emotion recognition method by fusing correlation features of speech-visual. Firstly, speech and visual features are extracted by two-dimensional convolutional neural network (2D-CNN) and three-dimensional convolutional neural network (3D-CNN), respectively. Secondly, the speech and visual features is processed by feature

We present an alternative inference framework for the Gaussian process-based extended object tracking (GPEOT) models. The method provides an approximate solution to the Bayesian filtering problem in GPEOT by relying on a new measurement update, which we derive using variational Bayes techniques. The resulting algorithm effectively computes approximate posterior densities of the kinematic and the extent

This letter considers the performance of the SPICE method in the context of line spectrum estimation by exploiting its equivalent square-root lasso (SR-LASSO) representation established in previous work. The inefficiencies of existing analyses are that the guarantees are not applicable to deterministic Fourier measurement matrices, and the studied range of the tuning parameter excludes the SR-LASSO

In this letter, we develop a coarray interpolation method for direction-of-arrival (DOA) estimation using the coprime electromagnetic vector-sensor (EMVS) array. Firstly, we derive the coarray signal model of the coprime EMVS array, which can be viewed as a combination of multiple polarization components. Subsequently, we fill in zero elements in each polarization component and recover the corresponding

This letter proposes a two-points non-iterative IpDFT algorithm for parameter estimation of the damped real-valued sinusoidal signal, which is named as the I2pNDFT. Both long- and short- range leakages are compensated to estimate the parameters accurately. Simulation results prove the effectiveness of our methodmagenta, compared with other IpDFT methods.

Temporal action localization aims at discovering action instances in untrimmed videos, where RGB and flow are two widely used feature modalities. Specifically, RGB chiefly reveals appearance and flow mainly depicts motion. Given RGB and flow features, previous methods employ the early fusion or late fusion paradigm to mine the complementarity between them. By concatenating raw RGB and flow features

High spectral resolution of hyperspectral images allows the detection and classification of materials in the observed images. However, existing research on hyperspectral detection mainly focuses on pixel-level study, partially due to the low spatial resolution in typical earth observation applications. With the development of imaging techniques, high-spatial-resolution hyperspectral data can be obtained