arXiv subjects

Hong Zhang

Publications and source records attributed to Hong Zhang.

At least 253 records · Page 14Linked to original sources

arXiv2021

Polarimetric Monocular Dense Mapping Using Relative Deep Depth Prior

This paper is concerned with polarimetric dense map reconstruction based on a polarization camera with the help of relative depth information as a prior. In general, polarization imaging is able to reveal information about surface normal such as azimuth and zenith angles, which can support the development of solutions to the problem of dense reconstruction, especially in texture-poor regions. However, polarimetric shape cues are ambiguous due to two types of polarized reflection (specular/diffuse). Although methods have been proposed to address this issue, they either are offline and therefore not practical in robotics applications, or use incomplete polarimetric cues, leading to sub-optimal performance. In this paper, we propose an online reconstruction method that uses full polarimetric cues available from the polarization camera. With our online method, we can propagate sparse depth values both along and perpendicular to iso-depth contours. Through comprehensive experiments on challenging image sequences, we demonstrate that our method is able to significantly improve the accuracy of the depthmap as well as increase its density, specially in regions of poor texture.

cs.CV↗

arXiv2021

A new type-II lepidocrocite-type TiO2/GaSe heterostructure: Electronic and optical properties, bandgap engineering, interaction with ultrafast laser pulses

Recently, van der Waals heterostructure has attracted interest both theoretically and experimentally for their potential applications in photoelectronic devices, photovoltaic devices, plasmonic devices and photocatalysis. Inspired by this, we design a lepidocrocite-type TiO2/GaSe heterostructure. Via first-principles simulations, we show that such a heterostructure is a direct bandgap semiconductor with a strong and broad optical absorption, ranging from visible light to UV region, exhibiting its potential application in photoelectronic and photovoltaic devices. With the planar-averaged electron density difference and Bader charge analysis, the heterostructure shows a strong capacity of enhancing the charge redistribution especially at the interface, prolonging the lifetime of excitons, and hence improving photocatalytic performance. By applying biaxial strain and interlayer coupling, the heterostructure exhibits a direct-indirect bandgap transition and shows a potential for mechanical sensors due to the smooth and linear variation of bandgaps. Furthermore, our result indicates that a lower interlayer distance leads to a stronger charge redistribution. The calculation of irradiating ultrafast on the heterostructure further reveals a semiconductor-metal transition for the heterostructure. Moreover, we find an enhanced induced plasmonic current in the heterostructure under both x-polarized and z-polarized laser, which is beneficial to plasmonic devices designs. Our research provides valuable insight in applying the lepidocrocite-type TiO2/GaSe heterostructure in photoelectronic, photovoltaic, photocatalytic, mechanical sensing and plasmonic realms.

cond-mat.mtrl-sci↗

arXiv2020

CauchyCP: a powerful test under non-proportional hazards using Cauchy combination of change-point Cox regressions

Non-proportional hazards data are routinely encountered in randomized clinical trials. In such cases, classic Cox proportional hazards model can suffer from severe power loss, with difficulty in interpretation of the estimated hazard ratio since the treatment effect varies over time. We propose CauchyCP, an omnibus test of change-point Cox regression models, to overcome both challenges while detecting signals of non-proportional hazards patterns. Extensive simulation studies demonstrate that, compared to existing treatment comparison tests under non-proportional hazards, the proposed CauchyCP test 1) controls the type I error better at small $α$ levels ($< 0.01$); 2) increases the power of detecting time-varying effects; and 3) is more computationally efficient. The superior performance of CauchyCP is further illustrated using retrospective analyses of two randomized clinical trial datasets and a pharmacogenetic biomarker study dataset. The R package $\textit{CauchyCP}$ is publicly available on CRAN.

stat.ME↗

arXiv2020

Direct Depth Learning Network for Stereo Matching

Being a crucial task of autonomous driving, Stereo matching has made great progress in recent years. Existing stereo matching methods estimate disparity instead of depth. They treat the disparity errors as the evaluation metric of the depth estimation errors, since the depth can be calculated from the disparity according to the triangulation principle. However, we find that the error of the depth depends not only on the error of the disparity but also on the depth range of the points. Therefore, even if the disparity error is low, the depth error is still large, especially for the distant points. In this paper, a novel Direct Depth Learning Network (DDL-Net) is designed for stereo matching. DDL-Net consists of two stages: the Coarse Depth Estimation stage and the Adaptive-Grained Depth Refinement stage, which are all supervised by depth instead of disparity. Specifically, Coarse Depth Estimation stage uniformly samples the matching candidates according to depth range to construct cost volume and output coarse depth. Adaptive-Grained Depth Refinement stage performs further matching near the coarse depth to correct the imprecise matching and wrong matching. To make the Adaptive-Grained Depth Refinement stage robust to the coarse depth and adaptive to the depth range of the points, the Granularity Uncertainty is introduced to Adaptive-Grained Depth Refinement stage. Granularity Uncertainty adjusts the matching range and selects the candidates' features according to coarse prediction confidence and depth range. We verify the performance of DDL-Net on SceneFlow dataset and DrivingStereo dataset by different depth metrics. Results show that DDL-Net achieves an average improvement of 25% on the SceneFlow dataset and $12\%$ on the DrivingStereo dataset comparing the classical methods. More importantly, we achieve state-of-the-art accuracy at a large distance.

cs.CV↗

arXiv2020

Full Matching on Low Resolution for Disparity Estimation

A Multistage Full Matching disparity estimation scheme (MFM) is proposed in this work. We demonstrate that decouple all similarity scores directly from the low-resolution 4D volume step by step instead of estimating low-resolution 3D cost volume through focusing on optimizing the low-resolution 4D volume iteratively leads to more accurate disparity. To this end, we first propose to decompose the full matching task into multiple stages of the cost aggregation module. Specifically, we decompose the high-resolution predicted results into multiple groups, and every stage of the newly designed cost aggregation module learns only to estimate the results for a group of points. This alleviates the problem of feature internal competitive when learning similarity scores of all candidates from one low-resolution 4D volume output from one stage. Then, we propose the strategy of \emph{Stages Mutual Aid}, which takes advantage of the relationship of multiple stages to boost similarity scores estimation of each stage, to solve the unbalanced prediction of multiple stages caused by serial multistage framework. Experiment results demonstrate that the proposed method achieves more accurate disparity estimation results and outperforms state-of-the-art methods on Scene Flow, KITTI 2012 and KITTI 2015 datasets.

cs.CV↗

arXiv2020

Forecasting confirmed cases of the COVID-19 pandemic with a migration-based epidemiological model

The unprecedented coronavirus disease 2019 (COVID-19) pandemic is still a worldwide threat to human life since its invasion into the daily lives of the public in the first several months of 2020. Predicting the size of confirmed cases is important for countries and communities to make proper prevention and control policies so as to effectively curb the spread of COVID-19. Different from the 2003 SARS epidemic and the worldwide 2009 H1N1 influenza pandemic, COVID-19 has unique epidemiological characteristics in its infectious and recovered compartments. This drives us to formulate a new infectious dynamic model for forecasting the COVID-19 pandemic within the human mobility network, named the SaucIR-model in the sense that the new compartmental model extends the benchmark SIR model by dividing the flow of people in the infected state into asymptomatic, pathologically infected but unconfirmed, and confirmed. Furthermore, we employ dynamic modeling of population flow in the model in order that spatial effects can be incorporated effectively. We forecast the spread of accumulated confirmed cases in some provinces of mainland China and other countries that experienced severe infection during the time period from late February to early May 2020. The novelty of incorporating the geographic spread of the pandemic leads to a surprisingly good agreement with published confirmed case reports. The numerical analysis validates the high degree of predictability of our proposed SaucIR model compared to existing resemblance. The proposed forecasting SaucIR model is implemented in Python. A web-based application is also developed by Dash (under construction).

stat.AP↗

arXiv2020

To Filter Prune, or to Layer Prune, That Is The Question

Recent advances in pruning of neural networks have made it possible to remove a large number of filters or weights without any perceptible drop in accuracy. The number of parameters and that of FLOPs are usually the reported metrics to measure the quality of the pruned models. However, the gain in speed for these pruned models is often overlooked in the literature due to the complex nature of latency measurements. In this paper, we show the limitation of filter pruning methods in terms of latency reduction and propose LayerPrune framework. LayerPrune presents a set of layer pruning methods based on different criteria that achieve higher latency reduction than filter pruning methods on similar accuracy. The advantage of layer pruning over filter pruning in terms of latency reduction is a result of the fact that the former is not constrained by the original model's depth and thus allows for a larger range of latency reduction. For each filter pruning method we examined, we use the same filter importance criterion to calculate a per-layer importance score in one-shot. We then prune the least important layers and fine-tune the shallower model which obtains comparable or better accuracy than its filter-based pruning counterpart. This one-shot process allows to remove layers from single path networks like VGG before fine-tuning, unlike in iterative filter pruning, a minimum number of filters per layer is required to allow for data flow which constraint the search space. To the best of our knowledge, we are the first to examine the effect of pruning methods on latency metric instead of FLOPs for multiple networks, datasets and hardware targets. LayerPrune also outperforms handcrafted architectures such as Shufflenet, MobileNet, MNASNet and ResNet18 by 7.3%, 4.6%, 2.8% and 0.5% respectively on similar latency budget on ImageNet dataset.

cs.CV↗

arXiv2020

Object-aware Feature Aggregation for Video Object Detection

We present an Object-aware Feature Aggregation (OFA) module for video object detection (VID). Our approach is motivated by the intriguing property that video-level object-aware knowledge can be employed as a powerful semantic prior to help object recognition. As a consequence, augmenting features with such prior knowledge can effectively improve the classification and localization performance. To make features get access to more content about the whole video, we first capture the object-aware knowledge of proposals and incorporate such knowledge with the well-established pair-wise contexts. With extensive experimental results on the ImageNet VID dataset, our approach demonstrates the effectiveness of object-aware knowledge with the superior performance of 83.93% and 86.09% mAP with ResNet-101 and ResNeXt-101, respectively. When further equipped with Sequence DIoU NMS, we obtain the best-reported mAP of 85.07% and 86.88% upon the paper submitted. The code to reproduce our results will be released after acceptance.

cs.CV↗

arXiv2020

Experimental and numerical analysis of cyclic deformation and fatigue behavior of a Mg-RE alloy

Strain-controlled fatigue of Mg-1Mn-0.5Nd (wt.\%) alloy were studied by experiment and simulation. The microstructure was made up of a dispersion of strongly texture grains (13\%) embedded in a matrix of grains with a random texture. The cyclic stress-strain curves showed limited tension-compression anisotropy because of the limited texture. Cyclic hardening under compression and cyclic softening under tension occurred due to the presence of twinning. Moreover, the twin volume fraction of the broken samples depended on whether the sample was broken in tension or compression, indicating that twining-detwinning occurs during the whole fatigue life. The mechanical response of the polycrystalline alloy was simulated by means of computational homogenization. The behavior of the Mg grains was modelled using a phenomenological crystal plasticity model that accounted for basal, prismatic and pyramidal slip (including isotropic and kinematic hardening) as well as twining and detwinning. The model parameters were calibrated from the cyclic stress-strain curves at different cyclic strain amplitudes. Numerical simulations were used to understand the dominant deformation mechanisms and to predict the fatigue life by means of a fatigue indicator parameter based on the accumulated plastic shear strain in each fatigue cycle.

cond-mat.mtrl-sci↗

arXiv2020

An efficient and accurate approximation to the distribution of quadratic forms of Gaussian variables

In computational and applied statistics, it is of great interest to get fast and accurate calculation for the distributions of the quadratic forms of Gaussian random variables. This paper presents a novel approximation strategy that contains two developments. First, we propose a faster numerical procedure in computing the moments of the quadratic forms. Second, we establish a general moment-matching framework for distribution approximation, which covers existing approximation methods for the distributions of the quadratic forms of Gaussian variables. Under this framework, a novel moment-ratio method (MR) is proposed to match the ratio of skewness and kurtosis based on the gamma distribution. Our extensive simulations show that 1) MR is almost as accurate as the exact distribution calculation and is much more efficient; 2) comparing with existing approximation methods, MR significantly improves the accuracy of approximating far right tail probabilities. The proposed method has wide applications. For example, it is a better choice than existing methods for facilitating hypothesis testing in big data analysis, where efficient and accurate calculation of very small $p$-values is desired. An R package Qapprox that implements related methods is available on CRAN.

stat.ME↗

arXiv2020

Regularized finite difference methods for the logarithmic Klein-Gordon equation

We propose and analyze two regularized finite difference methods for the logarithmic Klein-Gordon equation (LogKGE). Due to the blowup phenomena caused by the logarithmic nonlinearity of the LogKGE, it is difficult to construct numerical schemes and establish their error bounds. In order to avoid singularity, we present a regularized logarithmic Klein-Gordon equation (RLogKGE) with a small regularized parameter $0<\varepsilon\ll1$. Besides, two finite difference methods are adopted to solve the regularized logarithmic Klein-Gordon equation (RLogKGE) and rigorous error bounds are estimated in terms of the mesh size $h$, time step $τ$, and the small regularized parameter $\varepsilon$. Finally, numerical experiments are carried out to verify our error estimates of the two numerical methods and the convergence results from the LogKGE to the RLogKGE with the linear convergence order $O(\varepsilon)$.

math.AP↗

arXiv2020

Two regularized energy-preserving finite difference methods for the logarithmic Klein-Gordon equation

We present and analyze two regularized finite difference methods which preserve energy of the logarithmic Klein-Gordon equation (LogKGE). In order to avoid singularity caused by the logarithmic nonlinearity of the LogKGE, we propose a regularized logarithmic Klein-Gordon equation (RLogKGE) with a small regulation parameter $0<\varepsilon\ll1$ to approximate the LogKGE with the convergence order $O(\varepsilon)$. By adopting the energy method, the inverse inequality, and the cut-off technique of the nonlinearity to bound the numerical solution, the error bound $O(h^{2}+\frac{τ^{2}}{\varepsilon^{2}})$ of the two schemes with the mesh size $h$, the time step $τ$ and the parameter $\varepsilon$. Numerical results are reported to support our conclusions.

math.AP↗

arXiv2020

Implicit Semantic Data Augmentation for Deep Networks

In this paper, we propose a novel implicit semantic data augmentation (ISDA) approach to complement traditional augmentation techniques like flipping, translation or rotation. Our work is motivated by the intriguing property that deep networks are surprisingly good at linearizing features, such that certain directions in the deep feature space correspond to meaningful semantic transformations, e.g., adding sunglasses or changing backgrounds. As a consequence, translating training samples along many semantic directions in the feature space can effectively augment the dataset to improve generalization. To implement this idea effectively and efficiently, we first perform an online estimate of the covariance matrix of deep features for each class, which captures the intra-class semantic variations. Then random vectors are drawn from a zero-mean normal distribution with the estimated covariance to augment the training data in that class. Importantly, instead of augmenting the samples explicitly, we can directly minimize an upper bound of the expected cross-entropy (CE) loss on the augmented training set, leading to a highly efficient algorithm. In fact, we show that the proposed ISDA amounts to minimizing a novel robust CE loss, which adds negligible extra computational cost to a normal training procedure. Although being simple, ISDA consistently improves the generalization performance of popular deep models (ResNets and DenseNets) on a variety of datasets, e.g., CIFAR-10, CIFAR-100 and ImageNet. Code for reproducing our results is available at https://github.com/blackfeather-wang/ISDA-for-Deep-Networks.

cs.CV↗

arXiv2020

Material and debris transport patterns in Moreton Bay, Australia: The influence of Lagrangian coherent structures

Coastal tidal estuaries are vital to the exchange of energy and material between inland waters and the open ocean. Debris originating from the land and ocean enter this environment and are transported by currents (river outflow and tide), wind, waves and density gradients. Understanding and predicting the source and fate of such debris has considerable environmental, economic and visual importance. We show that this issue can be addressed using the Lagrangian coherent structures (LCS) technique which is highly robust to hydrodynamic model uncertainties. Here we present a comprehensive study showing the utility of this approach to describe the fate of floating material in a coastal tidal embayment. An example is given from Moreton Bay, a semi-enclosed subtropical embayment with high morphologic, ecological and economic significance to Southeast Queensland, Australia. Transport barriers visualised by the LCS create pathways and barriers for material transport in the embayment. It was found that the wind field modified both the rate attraction and location of the transport barriers. One of the key outcomes is the demonstration of the significant role of islands in partitioning the transport of material and mixing within the embayment. The distribution of the debris sources along the shoreline are explained by the relative location of the LCS to the shoreline. Therefore, extraction of LCS can help to predict sources and fate of anthropogenic marine debris and thus, serve as a useful way for effective management of vulnerable regions and marine protected areas.

physics.ao-ph↗

arXiv2020

Accurate $p$-Value Calculation for Generalized Fisher's Combination Tests Under Dependence

Combining dependent tests of significance has broad applications but the $p$-value calculation is challenging. Current moment-matching methods (e.g., Brown's approximation) for Fisher's combination test tend to significantly inflate the type I error rate at the level less than 0.05. It could lead to significant false discoveries in big data analyses. This paper provides several more accurate and computationally efficient $p$-value calculation methods for a general family of Fisher type statistics, referred as the GFisher. The GFisher covers Fisher's combination, Good's statistic, Lancaster's statistic, weighted Z-score combination, etc. It allows a flexible weighting scheme, as well as an omnibus procedure that automatically adapts proper weights and degrees of freedom to a given data. The new $p$-value calculation methods are based on novel ideas of moment-ratio matching and joint-distribution surrogating. Systematic simulations show that they are accurate under multivariate Gaussian, and robust under the generalized linear model and the multivariate $t$-distribution, down to at least $10^{-6}$ level. We illustrate the usefulness of the GFisher and the new $p$-value calculation methods in analyzing both simulated and real data of gene-based SNP-set association studies in genetics. Relevant computation has been implemented into R package $GFisher$.

stat.ME↗

arXiv2020

Gated Path Selection Network for Semantic Segmentation

Semantic segmentation is a challenging task that needs to handle large scale variations, deformations and different viewpoints. In this paper, we develop a novel network named Gated Path Selection Network (GPSNet), which aims to learn adaptive receptive fields. In GPSNet, we first design a two-dimensional multi-scale network - SuperNet, which densely incorporates features from growing receptive fields. To dynamically select desirable semantic context, a gate prediction module is further introduced. In contrast to previous works that focus on optimizing sample positions on the regular grids, GPSNet can adaptively capture free form dense semantic contexts. The derived adaptive receptive fields are data-dependent, and are flexible that can model different object geometric transformations. On two representative semantic segmentation datasets, i.e., Cityscapes, and ADE20K, we show that the proposed approach consistently outperforms previous methods and achieves competitive performance without bells and whistles.

cs.CV↗

arXiv2019

Large-scale Multi-modal Person Identification in Real Unconstrained Environments

Person identification (P-ID) under real unconstrained noisy environments is a huge challenge. In multiple-feature learning with Deep Convolutional Neural Networks (DCNNs) or Machine Learning method for large-scale person identification in the wild, the key is to design an appropriate strategy for decision layer fusion or feature layer fusion which can enhance discriminative power. It is necessary to extract different types of valid features and establish a reasonable framework to fuse different types of information. In traditional methods, different persons are identified based on single modal features to identify, such as face feature, audio feature, and head feature. These traditional methods cannot realize a highly accurate level of person identification in real unconstrained environments. The study aims to propose a fusion module to fuse multi-modal features for person identification in real unconstrained environments.

cs.CV↗

arXiv2019

Single-step Options for Adversary Driving

In this paper, we use reinforcement learning for safety driving in adversary settings. In our work, the knowledge in state-of-art planning methods is reused by single-step options whose action suggestions are compared in parallel with primitive actions. We show two advantages by doing so. First, training this reinforcement learning agent is easier and faster than training the primitive-action agent. Second, our new agent outperforms the primitive-action reinforcement learning agent, human testers as well as the state-of-art planning methods that our agent queries as skill options.

cs.AI↗