arXiv subjects

Hong Yang

Publications and source records attributed to Hong Yang.

At least 73 records · Page 4Linked to original sources

arXiv2022

LAB-Net: LAB Color-Space Oriented Lightweight Network for Shadow Removal

This paper focuses on the limitations of current over-parameterized shadow removal models. We present a novel lightweight deep neural network that processes shadow images in the LAB color space. The proposed network termed "LAB-Net", is motivated by the following three observations: First, the LAB color space can well separate the luminance information and color properties. Second, sequentially-stacked convolutional layers fail to take full use of features from different receptive fields. Third, non-shadow regions are important prior knowledge to diminish the drastic color difference between shadow and non-shadow regions. Consequently, we design our LAB-Net by involving a two-branch structure: L and AB branches. Thus the shadow-related luminance information can well be processed in the L branch, while the color property is well retained in the AB branch. In addition, each branch is composed of several Basic Blocks, local spatial attention modules (LSA), and convolutional filters. Each Basic Block consists of multiple parallelized dilated convolutions of divergent dilation rates to receive different receptive fields that are operated with distinct network widths to save model parameters and computational costs. Then, an enhanced channel attention module (ECA) is constructed to aggregate features from different receptive fields for better shadow removal. Finally, the LSA modules are further developed to fully use the prior information in non-shadow regions to cleanse the shadow regions. We perform extensive experiments on the both ISTD and SRD datasets. Experimental results show that our LAB-Net well outperforms state-of-the-art methods. Also, our model's parameters and computational costs are reduced by several orders of magnitude. Our code is available at https://github.com/ngrxmu/LAB-Net.

cs.CV↗

arXiv2022

Shadow-Aware Dynamic Convolution for Shadow Removal

With a wide range of shadows in many collected images, shadow removal has aroused increasing attention since uncontaminated images are of vital importance for many downstream multimedia tasks. Current methods consider the same convolution operations for both shadow and non-shadow regions while ignoring the large gap between the color mappings for the shadow region and the non-shadow region, leading to poor quality of reconstructed images and a heavy computation burden. To solve this problem, this paper introduces a novel plug-and-play Shadow-Aware Dynamic Convolution (SADC) module to decouple the interdependence between the shadow region and the non-shadow region. Inspired by the fact that the color mapping of the non-shadow region is easier to learn, our SADC processes the non-shadow region with a lightweight convolution module in a computationally cheap manner and recovers the shadow region with a more complicated convolution module to ensure the quality of image reconstruction. Given that the non-shadow region often contains more background color information, we further develop a novel intra-convolution distillation loss to strengthen the information flow from the non-shadow region to the shadow region. Extensive experiments on the ISTD and SRD datasets show our method achieves better performance in shadow removal over many state-of-the-arts. Our code is available at https://github.com/xuyimin0926/SADC.

cs.CV↗

arXiv2022

Experimental demonstration of phase-matching and Sagnac effect in a millimeter-scale wedged resonator gyroscope

The highly efficient coupling of light from conventional optical components to optical mode volumes lies in the heart of chip-based micro-devices, which is determined by the phase-matching between propagation constants of fiber taper and the whispering-gallery-mode (WGM) of the resonator. Optical gyroscopes, typically realized as fiber-optic gyroscopes and ring-laser gyroscopes, have been the mainstay in diverse applications such as positioning and inertial sensing. Here, the phase-matching is theoretically analyzed and experimentally verified. We observe Sagnac effect in a millimeter-scale wedged resonator gyroscope which has attracted considerable attention and been rapidly promoted in recent years. We demonstrate a bidirectional pump and probe scheme, which directly measures the frequency beat caused by the Sagnac effect. We establish the linear response between the detected beat frequency and the rotation velocity. The clockwise and counterclockwise rotation can also be distinguished according to the value of the frequency beat. The experimental results verify the feasibility of developing gyroscope in WGM resonator system and pave the way for future development.

physics.optics↗

arXiv2022

Symmetry-protected quantization of complex Berry phases in non-Hermitian many-body systems

We investigate the quantization of the complex-valued Berry phases in non-Hermitian quantum systems with certain generalized symmetries. In Hermitian quantum systems, the real-valued Berry phase is known to be quantized in the presence of certain symmetries, and this quantized Berry phase can be regarded as a topological order parameter for gapped quantum systems. In this paper, on the other hand, we establish that the complex Berry phase is also quantized in the systems described by a family of non-Hermitian Hamiltonians. Let $H(θ)$ be a non-Hermitian Hamiltonian parameterized by $θ$. Suppose that there exists a unitary and Hermitian operator $P$ such that $PH(θ)P = H(-θ)$ or $PH(θ)P = H^\dagger(-θ)$. We prove that in the former case, the complex Berry phase $γ$ is $\mathbb{Z}_2$-quantized, while in the latter, only the real part of $γ$ is $\mathbb{Z}_2$-quantized. The operator $P$ can be viewed as a generalized symmetry for $H(θ)$, and in practice, $P$ can be, for example, a spatial inversion. We also argue that this quantized complex Berry phase is capable of classifying non-Hermitian topological phases, and we demonstrate this in some one-dimensional strongly correlated systems.

cond-mat.str-el↗

arXiv2022

Predictive Maintenance for General Aviation Using Convolutional Transformers

Predictive maintenance systems have the potential to significantly reduce costs for maintaining aircraft fleets as well as provide improved safety by detecting maintenance issues before they come severe. However, the development of such systems has been limited due to a lack of publicly labeled multivariate time series (MTS) sensor data. MTS classification has advanced greatly over the past decade, but there is a lack of sufficiently challenging benchmarks for new methods. This work introduces the NGAFID Maintenance Classification (NGAFID-MC) dataset as a novel benchmark in terms of difficulty, number of samples, and sequence length. NGAFID-MC consists of over 7,500 labeled flights, representing over 11,500 hours of per second flight data recorder readings of 23 sensor parameters. Using this benchmark, we demonstrate that Recurrent Neural Network (RNN) methods are not well suited for capturing temporally distant relationships and propose a new architecture called Convolutional Multiheaded Self Attention (Conv-MHSA) that achieves greater classification performance at greater computational efficiency. We also demonstrate that image inspired augmentations of cutout, mixup, and cutmix, can be used to reduce overfitting and improve generalization in MTS classification. Our best trained models have been incorporated back into the NGAFID to allow users to potentially detect flights that require maintenance as well as provide feedback to further expand and refine the NGAFID-MC dataset.

cs.LG↗

arXiv2022

Robust Augmentation for Multivariate Time Series Classification

Neural networks are capable of learning powerful representations of data, but they are susceptible to overfitting due to the number of parameters. This is particularly challenging in the domain of time series classification, where datasets may contain fewer than 100 training examples. In this paper, we show that the simple methods of cutout, cutmix, mixup, and window warp improve the robustness and overall performance in a statistically significant way for convolutional, recurrent, and self-attention based architectures for time series classification. We evaluate these methods on 26 datasets from the University of East Anglia Multivariate Time Series Classification (UEA MTSC) archive and analyze how these methods perform on different types of time series data.. We show that the InceptionTime network with augmentation improves accuracy by 1% to 45% in 18 different datasets compared to without augmentation. We also show that augmentation improves accuracy for recurrent and self attention based architectures.

cs.LG↗

arXiv2022

Parameter-free quantum hydrodynamic theory for plasmonics: Electron density-dependent damping rate and diffusion coefficient

Plasmonics is a rapid growing field, which has enabled both fundamental science and inventions of various quantum optoelectronic devices. An accurate and efficient method to calculate the optical response of metallic structures with feature size in the nanoscale plays an important role. Quantum hydrodynamic theory (QHT) provides an efficient description of the free-electron gas, where quantum effects of nonlocality and spill-out are taken into account. In this work, we introduce a general QHT that includes diffusion to account for the broadening, which is a key problem in practical applications of surface plasmon. We will introduce a density-dependent diffusion coefficient to give very accurate linewidth. It is a self-consistent method, in which both the ground and excited states are solved by using the same energy functional, with the kinetic energy described by the Thomas-Fermi and von Weizsäcker (vW) formalisms. In addition, our QHT method is stable by introduction of an electron density-dependent damping rate. For sodium nanosphere of various sizes, the plasmon energy and broadening by our QHT method are in excellent agreement with those by density functional theory and Kreibig formula. By applying our QHT method to sodium jellium nanorods, we clearly show that our method enables a parameter-free simulation, i.e. without resorting to any empirical parameter, such as size-dependent damping rate and diffusing coefficient. It is found that there exists a perfect linear relation between the resonance wavelength and aspect radio. The width decreases with increasing aspect ratio and height. The calculations show that our QHT method provides an explicit and unified way to account for size-dependent frequency shifts and broadening of arbitrarily shaped geometries. It is reliable and robust with great predicability, and hence provides a general and efficient platform to study plasmonics.

cond-mat.mes-hall↗

arXiv2021

Tunable partial polarization beam splitter and optomechanically induced Faraday effect

Polarization beam splitter (PBS) is a crucial photonic element to separately extract transverse-electric (TE) and transverse-magnetic (TM) polarizations from the propagating light fields. Here, we propose a concise, continuously tunable and all-optical partial PBS in the vector optomechanical system which contains two orthogonal polarized cavity modes with degenerate frequency. The results show that one can manipulate the polarization states of different output fields by tuning the polarization angle of the pumping field and the system function as partial PBS when the pump laser polarizes vertically or horizontally. As a significant application of the tunable PBS, we propose a scheme of implementing quantum walks in resonator arrays without the aid of other auxiliary systems. Furthermore, we investigate the optomechanically induced Faraday effect in the vector optomechanical system which enables arbitrary tailoring of the input lights and the behaviors of polarization angles of the output fields in the under couple, critical couple, and over couple regimes. Our findings prove the optomechanical system is a potential platform to manipulate the polarization states in multimode resonators and boost the process of applications related to polarization modulation.

physics.optics↗

arXiv2021

Cell-Free Massive MIMO with Low-Complexity Hybrid Beamforming

Cell-Free Massive Multiple-input Multiple-output (mMIMO) consists of many access points (APs) in a coverage area that jointly serve the users. These systems can significantly reduce the interference among the users compared to conventional MIMO networks and so enable higher data rates and a larger coverage area. However, Cell-Free mMIMO systems face multiple practical challenges such as the high complexity and power consumption of the APs' analog front-ends. Motivated by prior works, we address these issues by considering a low complexity hybrid beamforming framework at the APs in which each AP has a limited number of RF-chains to reduce power consumption, and the analog combiner is designed only using the large-scale statistics of the channel to reduce the system's complexity. We provide closed-form expressions for the signal to interference and noise ratio (SINR) of both uplink and downlink data transmission with accurate random matrix approximations. Also, based on the existing literature, we provide a power optimization algorithm that maximizes the minimum SINR of the users for uplink scenario. Through several simulations, we investigate the accuracy of the derived random matrix approximations, trade-off between the 95% outage data rate and the number of RF-chains, and the impact of power optimization. We observe that the derived approximations accurately follow the exact simulations and that in uplink scenario while using MMSE combiner, power optimization does not improve the performance much.

cs.IT↗

arXiv2021

Symmetry-protected Topological Phases in Spinful Bosons with a Flat Band

We theoretically demonstrate that interacting symmetry-protected topological (SPT) phases can be realized with ultracold spinful bosonic atoms loaded on the lattices which have a flat band at the bottom of the band structure. Ground states of such systems are not conventional Mott insulators in the sense that the ground states possess not only spin fluctuations but also non-negligible charge fluctuations. The SPT phases in such systems are determined by both spin and charge fluctuations at zero temperature. We find that the many-body ground states of such systems can be exactly obtained in some special cases, and these exact ground states turn out to serve as representative states of the SPT phases. As a concrete example, we demonstrate that spin-1 bosons on a sawtooth chain can be in an SPT phase protected by $\mathbb{Z}_2 \times \mathbb{Z}_2$ spin rotation symmetry or time-reversal symmetry, and this SPT phase is a result of spin fluctuations. We also show that spin-3 bosons on a kagome lattice can be in an SPT phase protected by $D_2$ point group symmetry, but this SPT phase is, however, a result of charge fluctuations.

cond-mat.quant-gas↗

arXiv2021

Scalable higher-order exceptional surface with passive resonators

The sensitivity of perturbation sensing can be effectively enhanced with higher-order exceptional points due to the nonlinear response to frequency splitting. However, the experimental implementation is challenging since all the parameters need to be precisely prepared. The emergence of exceptional surface (ES) improves the robustness of the system to the external environment, while maintaining the same sensitivity. Here, we propose the first scalable protocol for realizing photonic high-order exceptional surface with passive resonators. By adding one or more additional passive resonators in the low-order ES photonic system, the 3- or arbitrary N-order ES is constructed and proved to be easily realized in experiment. We show that the sensitivity is enhanced and experimental demonstration is more resilent against the fabrication errors. The additional phase-modulation effect is also investigated.

physics.optics↗

arXiv2021

A Systematic Review of the Efforts and Hindrances of Modeling and Simulation of CAR T-cell Therapy

Chimeric Antigen Receptor (CAR) T-cell therapy is an immunotherapy that has recently become highly instrumental in the fight against life-threatening diseases. A variety of modeling and computational simulation efforts have addressed different aspects of CAR T therapy, including T-cell activation, T- and malignant cell population dynamics, therapeutic cost-effectiveness strategies, and patient survival analyses. In this article, we present a systematic review of those efforts, including mathematical, statistical, and stochastic models employing a wide range of algorithms, from differential equations to machine learning. To the best of our knowledge, this is the first review of all such models studying CAR T therapy. In this review, we provide a detailed summary of the strengths, limitations, methodology, data used, and data lacking in current published models. This information may help in designing and building better models for enhanced prediction and assessment of the benefit-risk balance associated with novel CAR T therapies, as well as with the data collection essential for building such models.

q-bio.QM↗

arXiv2021

Can Massive MIMO Support URLLC?

We investigate the feasibility of using Massive MIMO to support URLLC in both coherence interval based and 3GPP compliant pilot settings. We consider grant-free uplink transmission with MMSE receiver and adopt 3GPP channel models. In the coherence interval based pilot setting, by extensive system level simulations, we find that using a Massive MIMO base station with 128 antennas and MMSE receiver, URLLC requirements can be achieved in Urban Macro (UMa) Non-Line of Sight (NLoS) with orthogonal pilots and Neyman-Pearson detector. However, in the 3GPP compliant pilot setting, even by using the covariance matrix of Physical Resource Block (PRB) subcarriers for active UE detection and channel estimation as well as open-loop power control, we find that URLLC requirements are still challenging to achieve due to the insufficient pilot length and pilot symbol location regulations in a PRB.

cs.IT↗

arXiv2020

Optimally Supporting IoT with Cell-Free Massive MIMO

We study internet of things (IoT) systems supported by cell-free (CF) massive MIMO (mMIMO) with optimal linear channel estimation. For the uplink, we consider optimal linear MIMO receiver and obtain an uplink SINR approximation involving only large-scale fading coefficients using random matrix (RM) theory. Using this approximation we design several max-min power control algorithms that incorporate power and rate weighting coefficients to achieve a target rate with high energy efficiency. For the downlink, we consider maximum ratio (MR) beamforming. Instead of solving a complex quasi-concave problem for downlink power control, we employ a neural network (NN) technique to obtain comparable power control with around 30 times reduction in computation time. For large networks we proposed a different NN based power control algorithm. This algorithm is sub-optimal, but its big advantage is that it is scalable.

cs.IT↗

arXiv2020

A Scalable and Energy Efficient IoT System Supported by Cell-Free Massive MIMO

An IoT (Internet of things) system supports a massive number of IoT devices wirelessly. We show how to use Cell-Free Massive MIMO (multiple-input and multiple-output) to provide a scalable and energy efficient IoT system. We employ optimal linear estimation with random pilots to acquire CSI (channel state information) for MIMO precoding and decoding. In the uplink, we employ optimal linear decoder and utilize RM (random matrix) theory to obtain two accurate SINR (signal-to-interference plus noise ratio) approximations involving only large-scale fading coefficients. We derive several max-min type power control algorithms based on both exact SINR expression and RM approximations. Next, we consider the power control problem for downlink (DL) transmission. To avoid solving a time-consuming quasi-concave problem that requires repeat tests for the feasibility of a SOCP (second-order cone programming) problem, we develop a neural network (NN) aided power control algorithm that results in 30 times reduction in computation time. This power control algorithm leads to scalable Cell-Free Massive MIMO networks in which the amount of computations conducted by each AP does not depend on the number of network APs. Both UL and DL power control algorithms allow visibly improve the system spectral efficiency (SE) and, more importantly, lead to multi-fold improvements in Energy Efficiency (EE), which is crucial for IoT networks.

eess.SY↗

arXiv2020

e-TLD: Event-based Framework for Dynamic Object Tracking

This paper presents a long-term object tracking framework with a moving event camera under general tracking conditions. A first of its kind for these revolutionary cameras, the tracking framework uses a discriminative representation for the object with online learning, and detects and re-tracks the object when it comes back into the field-of-view. One of the key novelties is the use of an event-based local sliding window technique that tracks reliably in scenes with cluttered and textured background. In addition, Bayesian bootstrapping is used to assist real-time processing and boost the discriminative power of the object representation. On the other hand, when the object re-enters the field-of-view of the camera, a data-driven, global sliding window detector locates the object for subsequent tracking. Extensive experiments demonstrate the ability of the proposed framework to track and detect arbitrary objects of various shapes and sizes, including dynamic objects such as a human. This is a significant improvement compared to earlier works that simply track objects as long as they are visible under simpler background settings. Using the ground truth locations for five different objects under three motion settings, namely translation, rotation and 6-DOF, quantitative measurement is reported for the event-based tracking framework with critical insights on various performance issues. Finally, real-time implementation in C++ highlights tracking ability under scale, rotation, view-point and occlusion scenarios in a lab setting.

cs.CV↗

arXiv2020

An Interactive Data Visualization and Analytics Tool to Evaluate Mobility and Sociability Trends During COVID-19

The COVID-19 outbreak has dramatically changed travel behavior in affected cities. The C2SMART research team has been investigating the impact of COVID-19 on mobility and sociability. New York City (NYC) and Seattle, two of the cities most affected by COVID-19 in the U.S. were included in our initial study. An all-in-one dashboard with data mining and cloud computing capabilities was developed for interactive data analytics and visualization to facilitate the understanding of the impact of the outbreak and corresponding policies such as social distancing on transportation systems. This platform is updated regularly and continues to evolve with the addition of new data, impact metrics, and visualizations to assist public and decision-makers to make informed decisions. This paper presents the architecture of the COVID related mobility data dashboard and preliminary mobility and sociability metrics for NYC and Seattle.

cs.HC↗

arXiv2020

Cell-Free Massive MIMO with Nonorthogonal Pilots for Internet of Things

We consider Internet of Things (IoT) organized on the principles of cell-free massive MIMO. Since the number of things is very large, orthogonal pilots cannot be assigned to all of them even if the things are stationary. This results in an unavoidable pilot contamination problem, worsened by the fact that, for IoT, since the things are operating at very low transmit power. To mitigate this problem and achieve a high throughput, we use cell-free systems with optimal linear minimum mean squared error (LMMSE) channel estimation, while traditionally simple suboptimal estimators have been used in such systems. We further derive the analytical uplink and downlink signal-to-interference-plus-noise ratio (SINR) expressions for this scenario, which depends only on large scale fading coefficients. This allows us to design new power control algorithms that require only infrequent transmit power adaptation. Simulation results show a 40% improvement in uplink and downlink throughputs and 95% in energy efficiency over existing cell-free wireless systems and at least a three-fold uplink improvement over known IoT systems based on small-cell systems.

eess.SP↗