arXiv subjects

Wenzhe Liu

Publications and source records attributed to Wenzhe Liu.

32 records · Page 2Linked to original sources

arXiv2022

Low-latency Monaural Speech Enhancement with Deep Filter-bank Equalizer

It is highly desirable that speech enhancement algorithms can achieve good performance while keeping low latency for many applications, such as digital hearing aids, acoustically transparent hearing devices, and public address systems. To improve the performance of traditional low-latency speech enhancement algorithms, a deep filter-bank equalizer (FBE) framework was proposed, which integrated a deep learning-based subband noise reduction network with a deep learning-based shortened digital filter mapping network. In the first network, a deep learning model was trained with a controllable small frame shift to satisfy the low-latency demand, i.e., $\le$ 4 ms, so as to obtain (complex) subband gains, which could be regarded as an adaptive digital filter in each frame. In the second network, to reduce the latency, this adaptive digital filter was implicitly shortened by a deep learning-based framework, and was then applied to noisy speech to reconstruct the enhanced speech without the overlap-add method. Experimental results on the WSJ0-SI84 corpus indicated that the proposed deep FBE with only 4-ms latency achieved much better performance than traditional low-latency speech enhancement algorithms in terms of the indices such as PESQ, STOI, and the amount of noise reduction.

eess.AS↗

arXiv2022

A Neural Beam Filter for Real-time Multi-channel Speech Enhancement

Most deep learning-based multi-channel speech enhancement methods focus on designing a set of beamforming coefficients to directly filter the low signal-to-noise ratio signals received by microphones, which hinders the performance of these approaches. To handle these problems, this paper designs a causal neural beam filter that fully exploits the spatial-spectral information in the beam domain. Specifically, multiple beams are designed to steer towards all directions using a parameterized super-directive beamformer in the first stage. After that, the neural spatial filter is learned by simultaneously modeling the spatial and spectral discriminability of the speech and the interference, so as to extract the desired speech coarsely in the second stage. Finally, to further suppress the interference components especially at low frequencies, a residual estimation module is adopted to refine the output of the second stage. Experimental results demonstrate that the proposed approach outperforms many state-of-the-art multi-channel methods on the generated multi-channel speech dataset based on the DNS-Challenge dataset.

cs.SD↗

arXiv2021

Embedding and Beamforming: All-neural Causal Beamformer for Multichannel Speech Enhancement

The spatial covariance matrix has been considered to be significant for beamformers. Standing upon the intersection of traditional beamformers and deep neural networks, we propose a causal neural beamformer paradigm called Embedding and Beamforming, and two core modules are designed accordingly, namely EM and BM. For EM, instead of estimating spatial covariance matrix explicitly, the 3-D embedding tensor is learned with the network, where both spectral and spatial discriminative information can be represented. For BM, a network is directly leveraged to derive the beamforming weights so as to implement filter-and-sum operation. To further improve the speech quality, a post-processing module is introduced to further suppress the residual noise. Based on the DNS-Challenge dataset, we conduct the experiments for multichannel speech enhancement and the results show that the proposed system outperforms previous advanced baselines by a large margin in multiple evaluation metrics.

cs.SD↗

arXiv2021

Pixel Difference Networks for Efficient Edge Detection

Recently, deep Convolutional Neural Networks (CNNs) can achieve human-level performance in edge detection with the rich and abstract edge representation capacities. However, the high performance of CNN based edge detection is achieved with a large pretrained CNN backbone, which is memory and energy consuming. In addition, it is surprising that the previous wisdom from the traditional edge detectors, such as Canny, Sobel, and LBP are rarely investigated in the rapid-developing deep learning era. To address these issues, we propose a simple, lightweight yet effective architecture named Pixel Difference Network (PiDiNet) for efficient edge detection. Extensive experiments on BSDS500, NYUD, and Multicue are provided to demonstrate its effectiveness, and its high training and inference efficiency. Surprisingly, when training from scratch with only the BSDS500 and VOC datasets, PiDiNet can surpass the recorded result of human perception (0.807 vs. 0.803 in ODS F-measure) on the BSDS500 dataset with 100 FPS and less than 1M parameters. A faster version of PiDiNet with less than 0.1M parameters can still achieve comparable performance among state of the arts with 200 FPS. Results on the NYUD and Multicue datasets show similar observations. The codes are available at https://github.com/zhuoinoulu/pidinet.

cs.CV↗

arXiv2021

Locality Relationship Constrained Multi-view Clustering Framework

In most practical applications, it's common to utilize multiple features from different views to represent one object. Among these works, multi-view subspace-based clustering has gained extensive attention from many researchers, which aims to provide clustering solutions to multi-view data. However, most existing methods fail to take full use of the locality geometric structure and similarity relationship among samples under the multi-view scenario. To solve these issues, we propose a novel multi-view learning method with locality relationship constraint to explore the problem of multi-view clustering, called Locality Relationship Constrained Multi-view Clustering Framework (LRC-MCF). LRC-MCF aims to explore the diversity, geometric, consensus and complementary information among different views, by capturing the locality relationship information and the common similarity relationships among multiple views. Moreover, LRC-MCF takes sufficient consideration to weights of different views in finding the common-view locality structure and straightforwardly produce the final clusters. To effectually reduce the redundancy of the learned representations, the low-rank constraint on the common similarity matrix is considered additionally. To solve the minimization problem of LRC-MCF, an Alternating Direction Minimization (ADM) method is provided to iteratively calculate all variables LRC-MCF. Extensive experimental results on seven benchmark multi-view datasets validate the effectiveness of the LRC-MCF method.

cs.CV↗

arXiv2021

A Simultaneous Denoising and Dereverberation Framework with Target Decoupling

Background noise and room reverberation are regarded as two major factors to degrade the subjective speech quality. In this paper, we propose an integrated framework to address simultaneous denoising and dereverberation under complicated scenario environments. It adopts a chain optimization strategy and designs four sub-stages accordingly. In the first two stages, we decouple the multi-task learning w.r.t. complex spectrum into magnitude and phase, and only implement noise and reverberation removal in the magnitude domain. Based on the estimated priors above, we further polish the spectrum in the third stage, where both magnitude and phase information are explicitly repaired with the residual learning. Due to the data mismatch and nonlinear effect of DNNs, the residual noise often exists in the DNN-processed spectrum. To resolve the problem, we adopt a light-weight algorithm as the post-processing module to capture and suppress the residual noise in the non-active regions. In the Interspeech 2021 Deep Noise Suppression (DNS) Challenge, our submitted system ranked top-1 for the real-time track in terms of Mean Opinion Score (MOS) with ITU-T P.835 framework

cs.SD↗

arXiv2021

ICASSP 2021 Deep Noise Suppression Challenge: Decoupling Magnitude and Phase Optimization with a Two-Stage Deep Network

It remains a tough challenge to recover the speech signals contaminated by various noises under real acoustic environments. To this end, we propose a novel system for denoising in the complicated applications, which is mainly comprised of two pipelines, namely a two-stage network and a post-processing module. The first pipeline is proposed to decouple the optimization problem w:r:t: magnitude and phase, i.e., only the magnitude is estimated in the first stage and both of them are further refined in the second stage. The second pipeline aims to further suppress the remaining unnatural distorted noise, which is demonstrated to sufficiently improve the subjective quality. In the ICASSP 2021 Deep Noise Suppression (DNS) Challenge, our submitted system ranked top-1 for the real-time track 1 in terms of Mean Opinion Score (MOS) with ITU-T P.808 framework.

cs.SD↗

arXiv2020

Topological Polarization Singularities in Metaphotonics

Polarization singularities of vectorial electromagnetic fields locate at the positions (such as points, lines, or surfaces) where properties of polarization ellipses are not defined. They are manifested as circular and linear polarization, for which respectively the semi-major axes and normal vectors of polarization ellipses become indefinite. First observed in conical diffraction in the 1830s, the field of polarization singularities has been systematically reshaped and deepened by many pioneers of wave optics. Together with other exotic phenomena such as non-Hermiticity and topology, polarization singularities have been introduced into the vibrant field of nanophotonics, rendering unprecedented flexibilities for manipulations of light-matter interactions at the nanoscale. Here we review the recent results on the generation and observation of polarization singularities in metaphotonics. We start with the discussion of polarization singularities in the Mie theory, where both electric and magnetic multipoles are explored from perspectives of local and global polarization properties. We then proceed with the discussion of various photonic-crystal structures, for which both near- and far-field patterns manifest diverse polarization singularities characterized by the integer Poincare or more general half-integer Hopf indices (topological charges). Next, we review the most recent studies of conversions from polarization to phase singularities in scalar wave optics, demonstrating how bound states in the continuum can be exploited to generate directly optical vortices of various charges. Throughout our paper, we discuss and highlight several fundamental concepts and demonstrate their close connections and special links to metaphotonics. We believe polarization singularities can provide novel perspectives for light-matter manipulation for both fundamental studies and their practical applications.

physics.optics↗

arXiv2020

Routing of valley photons in a WS2 monolayer via delocalized Bloch modes of in-plane inversion-symmetry broken photonic crystal slabs

The valleys of two-dimensional transition metal dichalcogenides (TMDCs) offer a new degree of freedom for information processing. To take advantage of this valley degree of freedom, on one hand, it is feasible to control valleys by utilizing different external stimuli like optical and electric fields. On the other hand, nanostructures are also used to separate the valleys by near field coupling. However, for both above methods, either required low-temperature environment or low degree of coherence properties limit their further applications. Here, we demonstrate all-dielectric photonic crystal (PhC) slabs without in-plane inversion symmetry (C2 symmetry) could separate and route valley photons in a WS2 monolayer at room temperature. Coupling with circularly polarized photonic Bloch modes of such PhC slabs, valley photons emitted by a WS2 monolayer are routed directionally and efficiently separated in the far field. In addition, the far-field emission is directionally enhanced and with long-distance spatial coherence property.

physics.optics↗

arXiv2019

Generating optical vortex beams by momentum-space polarization vortices centered at bound states in the continuum

An optical vortex (OV) is a beam with spiral wave front and screw phase dislocation. This kind of beams is attracting rising interest in various fields. Here we theoretically proposed and experimentally realized a novel but easy approach to generate optical vortices. We leverage the inherent topological vortex structures of polarization around bound states in the continuum (BIC) in the momentum space of two dimensional periodic structures, e.g. photonic crystal slabs, to induce Pancharatnam-Berry phases to the beams. This new class of OV generators operates in the momentum space, meaning that there is no real-space center of structure. Thus, not only the fabrication but also the practical alignment would be greatly simplified. Any even order of OV, which is actually a quasi-non-diffractive high-order quasi-Bessel beam, at any desired working wavelength could be achieved in principle. The proposed approach expands the application of bound states in the continuum and topological photonics.

physics.optics↗

arXiv2019

Circularly polarized states spawning from bound states in the continuum

Bound states in the continuum in periodic photonic systems like photonic crystal slabs are proved to be accompanied by vortex polarization singularities on the photonic bands in the momentum space. The winding structures of polarization states not only widen the field of topological physics but also show great potential that such systems could be applied in polarization manipulating. In this work, we report the phenomenon that by in-plane inversion ($C_2$) symmetry breaking, pairs of circularly polarized states could spawn from the eliminated Bound states in the continuum. Along with the appearance of the circularly polarized states as the two poles of the Poincaré sphere together with linearly polarized states covering the equator, full coverage on the Poincaré sphere could be realized. As an application, ellipticity modulation of linear polarization is demonstrated in the visible frequency range. This phenomenon provides new degree of freedom in modulating polarization.

physics.optics↗

arXiv2018

Observation of optical states below light cone with compound lattices

For optical systems, states inside the light cone could be detected by far-field measurement, while those below the light cone are not detectable by far-field measurement. A new method for far-field detection has been developed for observing states below the light cone with compound lattices. The basic mechanism involved is that periodic weak scattering leads to band folding, making the states out of the light cone to occur inside. By using polarization-resolved momentum-space imaging spectroscopy the band structures and iso-frequency contours of plasmonic lattices with different dimensions and symmetries are experimentally mapped out, in good agreement with the simulation.

physics.optics↗

arXiv2017

Observing half and integer polarization vortices at band degeneracies

Far-field polarization vortices were recently found on singlet bands in the momentum-space of two-dimensional photonic lattices, also known as the dark states and bound states in continuum. Here, we theoretically proposed and experimentally verified the existence of the polarization vortices at the degenerate points of photonic dispersions, whose vortex cores can be radiative bright states. Half-charged vortices were generated from the Dirac points of π Berry phase and integer-charged vortices were generated from a quadratic degeneracy. Using a home-made polarization-resolved momentum-space imaging spectroscopy, we observed the complete evolution of the splitting from one quadratic point to a pair of Dirac cones by tracking the winding of the polarization vectors and the full spectrum of iso-frequency contours.

physics.optics↗

arXiv2017

Observation of optical vortices in momentum space

Vortex, the winding of a vector field in two dimensions, has its core the field singularity and its topological charge defined by the quantized winding angle of the vector field. Vortices are one of the most fundamental topological excitations in nature, widely known in hair whorls as the winding of hair strings, in fluid dynamics as the winding of velocities, in angular-momentum beams as the winding of phase angle and in superconductors and superfluids as the winding of order parameters. Nevertheless, vortices have hardly been observed other than those in the real space. Although band degeneracies, such as Dirac cones, can be viewed as momentum-space vortices in their mathematical structures, there lacks a well-defined physical observable whose winding number is an arbitrary signed integer. Here, we experimentally observed momentum-space vortices as the winding of far-field polarization vectors in the Brillouin zone (BZ) of periodic plasmonic structures. Using a home-made polarization-resolved momentum-space imaging spectroscopy, we completely map out the dispersion, lifetime and polarization of all radiative states at the visible wavelengths. The momentum space vortices were experimentally identified by their winding patterns in the polarization-resolved iso-frequency contours and their diverging radiative quality factors. Such polarization vortices can exist robustly on any periodic systems of vectorial fields, while they are not captured by the existing topological band theory developed for scaler fields. This work opens up a promising avenue for exploring topological photonics in the momentum space, studying bound states in continuum (BICs), as well as for rendering and steering vector beams and designing high-Q plasmonic resonances.

physics.optics↗