arXiv subjects

Jun Zhang

Publications and source records attributed to Jun Zhang.

At least 937 records · Page 52Linked to original sources

arXiv2019

Comparing the route-choice behavior of pedestrians around obstacles in a virtual experiment and a field study

Pedestrians often need to decide between different routes they can use to reach their intended destinations, both during emergencies and in their daily lives. This route-choice behavior is important in determining traffic management, evacuation efficiency and building design. Here, we use field observations and a virtual experiment to study the route choice behavior of pedestrians around obstacles delimiting exit routes and examine the influence of three factors, namely the local distance to route starting points and the pedestrian density and walking speeds along routes. Crucially, both field study and virtual experiment consider the same scenario which allows us to directly assess the validity of testing pedestrian behavior in virtual environments. We find that in both data sets the proportion of people who choose a closer exit route increases as the difference in distance between exit route starting points increases. Pedestrians' choices in our data also depend on pedestrian density along routes, with people preferring less used routes. Our results thus confirm previously established route choice mechanisms and we can predict over 74% of choices based on these factors. The qualitative agreement in results between the field study and the virtual experiment suggests that in simple route-choice scenarios, such as the one we investigate here, virtual experiments can be a valid experimental technique for studying pedestrian behavior. We therefore provide much-needed empirical support for the emerging paradigm of experiments in virtual environments.

physics.soc-ph↗

arXiv2019

Artificial Neural Network Based Modeling on Unidirectional and Bidirectional Pedestrian Flow at Straight Corridors

Pedestrian modeling is a good way to predict pedestrian movement and thus can be used for controlling pedestrian crowds and guiding evacuations in emergencies. In this paper, we propose a pedestrian movement model based on artificial neural network. In the model, the pedestrian velocity vectors are predicted with two sub models, Semicircular Forward Space Based submodel (SFSB-submodel) and Rectangular Forward Space Based submodel (RFSB-submodel), respectively. Both unidirectional and bidirectional pedestrian flow at straight corridors are investigated by comparing the simulation and the corresponding experimental results. It is shown that the pedestrian trajectories and the fundamental diagrams from the model are all consistent with that from experiments. And the typical lane-formation phenomena are observed in bidirectional flow simulation. In addition, to quantitatively evaluate the precision of the prediction, the mean destination error (MDE) and mean trajectory error (MTE) are defined and calculated to be approximately 0.2m and 0.12m in unidirectional flow scenario. In bidirectional flow, relative distance error (RDE) is about 0.15m. The findings indicate that the proposed model is reasonable and capable of simulating the unidirectional and bidirectional pedestrian flow illustrated in this paper.

physics.soc-ph↗

arXiv2019

Contrastive study on the single-file pedestrian movement of the elderly and other age groups

The worldwide population is aging and countries are facing ongoing challenges in improving the safety of elderly pedestrians. In this work, single-file movement of the elderly are experimentally compared with that of different age groups. The findings indicates that the age is not the only factor influencing the pedestrian dynamics but the heterogeneity of the crowd composition and the familiarity among neighboring pedestrians also have significant effects. The existence of three regimes in the relationship between headway and speed is confirmed. In the strong constrained regime, the slope of the relationship between headway and speed of the elderly is bigger than that of the young, which means that the elders are more sensitive to the spatial headway than the young when adapting the speeds. However, the difference of the slopes in the weakly constrained regime is small, which indicates a weak dependency between age and the adaption time. The elderly need longer headway during the transformation of the motion state. Besides, the 'active cease' behavior of pedestrians, which is explained with the least effort principle, is observed in the experiment. The findings offer empirical data of the elderly under high densities and can be useful for the improvement of the pedestrian modelling and the construction of elderly friendly facilities.

physics.soc-ph↗

arXiv2019

Experimental study on the elderly pedestrians passing through bottlenecks

The aging of population is a social phenomenon in the world. In this study, series of controlled experiments were performed to investigate the movement characteristics of elderly pedestrians passing through bottlenecks. Similar self-organized phenomena to that of the young like 'zipper effect' and 'lane formation' are observed. The highest local density that appears in front of the exit in the experiment is beyond 4 m-2. The time lapse between two consecutive pedestrians of the elderly is about 0.2 s longer than that of the young when the width is 0.5 m and 0.7 m. Besides, the difference decreases with the increasing exit width. A linear relation between flow and bottleneck width is obtained, while the flow of elders is lower than that of the young under the same width of exits. The waiting time of pedestrians can be divided into two stages when the bottleneck width is narrower than 0.8 m. The findings in this study will benefit the evacuation guidance as well as the facility design for elders.

physics.soc-ph↗

arXiv2019

Experimental study of pedestrian flow mixed with wheelchair users through funnel-shaped bottlenecks

With the increase of the elderly and disabled in the world, the characteristics of pedestrian flow mixed with wheelchair users has being paid more and more attention. In this study, experiments in funnel-shaped bottlenecks were performed to study the impact of bottleneck shape and the ratio of wheelchair users on the crowd dynamics. It is found that the increase of wheelchairs in the crowd obviously leads to worse moving efficiency and congestion from the escape time, time-space relationship and time headway. Under low mixing ratios (<2.35%), less congestion occurred in the 45° bottleneck among the four tested angles (0°, 15°, 30°, 45°). The average speeds of wheelchair users is the fastest in 45 ° bottleneck (0.310m/s) until the mixing ratio arrives at 7.05%. However, the advantage of the angle disappears when the mixing ratio gets higher. The findings in this study is meaningful for the guidance of pedestrian evacuation through bottlenecks with the presence of wheelchair users.

physics.soc-ph↗

arXiv2019

Persistence modules, symplectic Banach-Mazur distance and Riemannian metrics

We use persistence modules and their corresponding barcodes to quantitatively distinguish between different fiberwise star-shaped domains in the cotangent bundle of a fixed manifold. The distance between two fiberwise star-shaped domains is measured by a non-linear version of the classical Banach-Mazur distance, called symplectic Banach-Mazur distance and denoted by $d_{SBM}.$ The relevant persistence modules come from filtered symplectic homology and are stable with respect to $d_{SBM}.$ Our main focus is on the space of unit codisc bundles of orientable surfaces of positive genus, equipped with Riemannian metrics. We consider some questions about large scale-geometry of this space and in particular we give a construction of a quasi-isometric embedding of $(\mathbb{R}^n,|\cdot |_\infty)$ into this space for all $n\in \mathbb{N}.$ On the other hand, in the case of domains in $T^*S^2$, we can show that the corresponding metric space has infinite diameter. Finally, we discuss the existence of closed geodesics whose energies can be controlled.

math.SG↗

arXiv2019

Advances in Online Audio-Visual Meeting Transcription

This paper describes a system that generates speaker-annotated transcripts of meetings by using a microphone array and a 360-degree camera. The hallmark of the system is its ability to handle overlapped speech, which has been an unsolved problem in realistic settings for over a decade. We show that this problem can be addressed by using a continuous speech separation approach. In addition, we describe an online audio-visual speaker diarization method that leverages face tracking and identification, sound source localization, speaker identification, and, if available, prior speaker information for robustness to various real world challenges. All components are integrated in a meeting transcription framework called SRD, which stands for "separate, recognize, and diarize". Experimental results using recordings of natural meetings involving up to 11 attendees are reported. The continuous speech separation improves a word error rate (WER) by 16.1% compared with a highly tuned beamformer. When a complete list of meeting attendees is available, the discrepancy between WER and speaker-attributed WER is only 1.0%, indicating accurate word-to-speaker association. This increases marginally to 1.6% when 50% of the attendees are unknown to the system.

eess.AS↗

arXiv2019

Hybrid Beamforming for 5G and Beyond Millimeter-Wave Systems: A Holistic View

Millimeter-wave (mm-wave) communication is a key technology for future wireless networks. To combat significant path loss and exploit the abundant mm-wave spectrum, effective beamforming is crucial. Nevertheless, conventional fully digital beamforming techniques are inapplicable, as they demand a separate radio frequency (RF) chain for each antenna element, which is costly and consumes too much energy. Hybrid beamforming is a cost-effective alternative, which can significantly reduce the hardware cost and power consumption by employing a small number of RF chains. This paper presents a holistic view on hybrid beamforming for 5G and beyond mm-wave systems, based on a new taxonomy for different hardware structures. We take a pragmatic approach and compare different proposals from three key aspects: 1) hardware efficiency, i.e., the required hardware components; 2) computational efficiency of the associated beamforming algorithm; and 3) achievable spectral efficiency, a main performance indicator. Through systematic comparisons, the interplay and trade-off among these three design aspects are demonstrated, and promising candidates for hybrid beamforming in future wireless networks are identified.

eess.SP↗

arXiv2019

Photonic realization of quantum resetting

Contrary to the usual assumption of at least partial control of quantum dynamics, a surprising recent result proved that an arbitrary quantum state can be probabilistically reset to a state in the past by having it interact with probing systems in a consistent, but $uncontrolled$ way. We present a photonic implementation to achieve this resetting process, experimentally verifying that a state can be probabilistically reset to its past with a fidelity of $0.870\pm0.012$. We further demonstrate the preservation of an entangled state, which still violates a Bell inequality, after half of the entangled pair was reset. The ability to reset uncontrolled quantum states has implications in the foundations of quantum physics and applications in areas of quantum technology.

quant-ph↗

arXiv2019

A Study of Data Pre-processing Techniques for Imbalanced Biomedical Data Classification

Biomedical data are widely accepted in developing prediction models for identifying a specific tumor, drug discovery and classification of human cancers. However, previous studies usually focused on different classifiers, and overlook the class imbalance problem in real-world biomedical datasets. There are a lack of studies on evaluation of data pre-processing techniques, such as resampling and feature selection, on imbalanced biomedical data learning. The relationship between data pre-processing techniques and the data distributions has never been analysed in previous studies. This article mainly focuses on reviewing and evaluating some popular and recently developed resampling and feature selection methods for class imbalance learning. We analyse the effectiveness of each technique from data distribution perspective. Extensive experiments have been done based on five classifiers, four performance measures, eight learning techniques across twenty real-world datasets. Experimental results show that: (1) resampling and feature selection techniques exhibit better performance using support vector machine (SVM) classifier. However, resampling and Feature Selection techniques perform poorly when using C4.5 decision tree and Linear discriminant analysis classifiers; (2) for datasets with different distributions, techniques such as Random undersampling and Feature Selection perform better than other data pre-processing methods with T Location-Scale distribution when using SVM and KNN (K-nearest neighbours) classifiers. Random oversampling outperforms other methods on Negative Binomial distribution using Random Forest classifier with lower level of imbalance ratio; (3) Feature Selection outperforms other data pre-processing methods in most cases, thus, Feature Selection with SVM classifier is the best choice for imbalanced biomedical data learning.

q-bio.QM↗

arXiv2019

Breakdown of Raman Selection Rules By Fröhlich Interaction in Few-Layer WS$_2$

The Raman selection rules arise from the crystal symmetry and then determine the Raman activity and polarization of scattered phonon modes. However, these selection rules can be broken in resonant process due to the strong electron-phonon coupling effect. Here we reported the observation of breakdown of Raman selection rules in few-layer WS$_2$ by using resonant Raman scattering with dark A exciton. In this case, not only the infrared active modes and backscattering forbidden modes are observed, but the intensities of all observed phonon modes become strongest under paralleled-polarization and independent on the Raman tensors of phonons. We attributed this phenomenon to the interaction between dark A exciton and the scatted phonon, the so-called intraband Fröhlich interaction, where the Raman scattering possibility is totally determined by the symmetry of exciton rather than the phonons due to strong electron-phonon coupling. Our results not only can be used to easily detect the optical forbidden excitonic and phononic states but also provide a possible way to manipulate optical transitions between electronic levels.

cond-mat.mtrl-sci↗

arXiv2019

Client-Edge-Cloud Hierarchical Federated Learning

Federated Learning is a collaborative machine learning framework to train a deep learning model without accessing clients' private data. Previous works assume one central parameter server either at the cloud or at the edge. The cloud server can access more data but with excessive communication overhead and long latency, while the edge server enjoys more efficient communications with the clients. To combine their advantages, we propose a client-edge-cloud hierarchical Federated Learning system, supported with a HierFAVG algorithm that allows multiple edge servers to perform partial model aggregation. In this way, the model can be trained faster and better communication-computation trade-offs can be achieved. Convergence analysis is provided for HierFAVG and the effects of key parameters are also investigated, which lead to qualitative design guidelines. Empirical experiments verify the analysis and demonstrate the benefits of this hierarchical architecture in different data distribution scenarios. Particularly, it is shown that by introducing the intermediate edge servers, the model training time and the energy consumption of the end devices can be simultaneously reduced compared to cloud-based Federated Learning.

cs.NI↗

arXiv2019

Light Field Saliency Detection with Deep Convolutional Networks

Light field imaging presents an attractive alternative to RGB imaging because of the recording of the direction of the incoming light. The detection of salient regions in a light field image benefits from the additional modeling of angular patterns. For RGB imaging, methods using CNNs have achieved excellent results on a range of tasks, including saliency detection. However, it is not trivial to use CNN-based methods for saliency detection on light field images because these methods are not specifically designed for processing light field inputs. In addition, current light field datasets are not sufficiently large to train CNNs. To overcome these issues, we present a new Lytro Illum dataset, which contains 640 light fields and their corresponding ground-truth saliency maps. Compared to current light field saliency datasets [1], [2], our new dataset is larger, of higher quality, contains more variation and more types of light field inputs. This makes our dataset suitable for training deeper networks and benchmarking. Furthermore, we propose a novel end-to-end CNN-based framework for light field saliency detection. Specifically, we propose three novel MAC (Model Angular Changes) blocks to process light field micro-lens images. We systematically study the impact of different architecture variants and compare light field saliency with regular 2D saliency. Our extensive comparisons indicate that our novel network significantly outperforms state-of-the-art methods on the proposed dataset and has desired generalization abilities on other existing datasets.

cs.CV↗

arXiv2019

Research progress based on observations of the New Vacuum Solar Telescope

The purpose of this paper is to introduce the main scientific results made by the one-meter New Vacuum Solar Telescope (NVST), which was put into commission on 2010. NVST is one of the large aperture solar telescopes in the world, located on the shore of Fuxian lake of Yunnan province in China, aiming at serving solar physicists by providing them with high resolution photospheric and chromospheric observational data. Based on the data from NVST and complementary observations from space (e.g., Hinode, SDO and IRIS, etc), dozens of scientific papers have been published with a wide range of topics concentrating mainly on dynamics and activities of fine-scale magnetic structures and their roles in the eruptions of active-region filaments and flares. The achievements include dynamic characteristics of photospheric bright points, umbral dots, penumbral waves, and sunspot/light bridge oscillation, observational evidence of small-scale magnetic reconnection, and fine-scale dynamic structure of prominences. All these new results will shed light on the better understanding of solar eruptive activities. Data release, observation proposals, and future research subjects are introduced and discussed.

astro-ph.SR↗

arXiv2019

Observation of Fano resonance in photoluminescence of a two-dimensional magnetic semiconductor

Quantum interference gives rise to the asymmetric Fano resonance line shape when the final states of an electronic transition follows within a continuum of states and a discrete state, which has significant applications in optical switching and sensing. The resonant optical phenomena associated with Fano resonance have been observed by absorption spectra, Raman spectra, transmission spectra, etc., but have rarely been reported in photoluminescence (PL) spectroscopy. In this work, we performed spectroscopic studies on layered chromium thiophosphate (CrPS4), a promising ternary antiferromagnetic semiconductor with PL in near-infrared wavelength region and observed Fano resonance when CrPS4 experiences phase transition into the antiferromagnetic state below Néel temperature (38 K). The photoluminescence of the continuum states results from the d band transitions localized at Cr3+ ions, while the discrete state reaches saturation at high excitation power and can be enhanced by the external magnetic field, suggesting it is formed by an impurity level from extra atomic phosphorus. Our findings provide insights into the electronic transitions of CrPS4 and their connection to its intrinsic magnetic properties.

cond-mat.mtrl-sci↗

arXiv2019

Data-Importance Aware User Scheduling for Communication-Efficient Edge Machine Learning

With the prevalence of intelligent mobile applications, edge learning is emerging as a promising technology for powering fast intelligence acquisition for edge devices from distributed data generated at the network edge. One critical task of edge learning is to efficiently utilize the limited radio resource to acquire data samples for model training at an edge server. In this paper, we develop a novel user scheduling algorithm for data acquisition in edge learning, called (data) importance-aware scheduling. A key feature of this scheduling algorithm is that it takes into account the informativeness of data samples, besides communication reliability. Specifically, the scheduling decision is based on a data importance indicator (DII), elegantly incorporating two "important" metrics from communication and learning perspectives, i.e., the signal-to-noise ratio (SNR) and data uncertainty. We first derive an explicit expression for this indicator targeting the classic classifier of support vector machine (SVM), where the uncertainty of a data sample is measured by its distance to the decision boundary. Then, the result is extended to convolutional neural networks (CNN) by replacing the distance based uncertainty measure with the entropy. As demonstrated via experiments using real datasets, the proposed importance-aware scheduling can exploit the two-fold multi-user diversity, namely the diversity in both the multiuser channels and the distributed data samples. This leads to faster model convergence than the conventional scheduling schemes that exploit only a single type of diversity.

cs.NI↗

arXiv2019

Multi-grained Attention Networks for Single Image Super-Resolution

Deep Convolutional Neural Networks (CNN) have drawn great attention in image super-resolution (SR). Recently, visual attention mechanism, which exploits both of the feature importance and contextual cues, has been introduced to image SR and proves to be effective to improve CNN-based SR performance. In this paper, we make a thorough investigation on the attention mechanisms in a SR model and shed light on how simple and effective improvements on these ideas improve the state-of-the-arts. We further propose a unified approach called "multi-grained attention networks (MGAN)" which fully exploits the advantages of multi-scale and attention mechanisms in SR tasks. In our method, the importance of each neuron is computed according to its surrounding regions in a multi-grained fashion and then is used to adaptively re-scale the feature responses. More importantly, the "channel attention" and "spatial attention" strategies in previous methods can be essentially considered as two special cases of our method. We also introduce multi-scale dense connections to extract the image features at multiple scales and capture the features of different layers through dense skip connections. Ablation studies on benchmark datasets demonstrate the effectiveness of our method. In comparison with other state-of-the-art SR methods, our method shows the superiority in terms of both accuracy and model size.

eess.IV↗

arXiv2019

Non-imaging single-pixel sensing with optimized binary modulation

The conventional high-level sensing techniques require high-fidelity images as input to extract target features, which are produced by either complex imaging hardware or high-complexity reconstruction algorithms. In this letter, we propose single-pixel sensing (SPS) that performs high-level sensing directly from coupled measurements of a single-pixel detector, without the conventional image acquisition and reconstruction process. The technique consists of three steps including binary light modulation that can be physically implemented at $\sim$22kHz, single-pixel coupled detection owning wide working spectrum and high signal-to-noise ratio, and end-to-end deep-learning based sensing that reduces both hardware and software complexity. Besides, the binary modulation is trained and optimized together with the sensing network, which ensures least required measurements and optimal sensing accuracy. The effectiveness of SPS is demonstrated on the classification task of handwritten MNIST dataset, and 96.68% classification accuracy at $\sim$1kHz is achieved. The reported single-pixel sensing technique is a novel framework for highly efficient machine intelligence.

eess.IV↗