arXiv subjects

Liang Peng

Publications and source records attributed to Liang Peng.

At least 91 records · Page 5Linked to original sources

arXiv2021

OCM3D: Object-Centric Monocular 3D Object Detection

Image-only and pseudo-LiDAR representations are commonly used for monocular 3D object detection. However, methods based on them have shortcomings of either not well capturing the spatial relationships in neighbored image pixels or being hard to handle the noisy nature of the monocular pseudo-LiDAR point cloud. To overcome these issues, in this paper we propose a novel object-centric voxel representation tailored for monocular 3D object detection. Specifically, voxels are built on each object proposal, and their sizes are adaptively determined by the 3D spatial distribution of the points, allowing the noisy point cloud to be organized effectively within a voxel grid. This representation is proved to be able to locate the object in 3D space accurately. Furthermore, prior works would like to estimate the orientation via deep features extracted from an entire image or a noisy point cloud. By contrast, we argue that the local RoI information from the object image patch alone with a proper resizing scheme is a better input as it provides complete semantic clues meanwhile excludes irrelevant interferences. Besides, we decompose the confidence mechanism in monocular 3D object detection by considering the relationship between 3D objects and the associated 2D boxes. Evaluated on KITTI, our method outperforms state-of-the-art methods by a large margin. The code will be made publicly available soon.

cs.CV↗

arXiv2020

Geometry-based Occlusion-Aware Unsupervised Stereo Matching for Autonomous Driving

Recently, there are emerging many stereo matching methods for autonomous driving based on unsupervised learning. Most of them take advantage of reconstruction losses to remove dependency on disparity groundtruth. Occlusion handling is a challenging problem in stereo matching, especially for unsupervised methods. Previous unsupervised methods failed to take full advantage of geometry properties in occlusion handling. In this paper, we introduce an effective way to detect occlusion regions and propose a novel unsupervised training strategy to deal with occlusion that only uses the predicted left disparity map, by making use of its geometry features in an iterative way. In the training process, we regard the predicted left disparity map as pseudo groundtruth and infer occluded regions using geometry features. The resulting occlusion mask is then used in either training, post-processing, or both of them as guidance. Experiments show that our method could deal with the occlusion problem effectively and significantly outperforms the other unsupervised methods for stereo matching. Moreover, our occlusion-aware strategies can be extended to the other stereo methods conveniently and improve their performances.

cs.CV↗

arXiv2019

Transverse photon spin beyond interfaces

Photons possess spin degree of freedom, corresponding to clockwise and counter clockwise rotating direction of the fields. Photon spin plays an important role in various applications such as optical communications, information processing and sensing. In conventional isotropic media, photon spin is aligned with the propagation direction of light, obeying spin momentum locking. Interestingly, at certain interfaces, the surface waves decaying away from the interface possess a photon spin transverse to its propagation, opening exciting opportunities for observation of spin dependent unidirectional excitation in confined systems. Here we propose and realize transverse photon spin (T-spin) in the interior of a bulk medium, without relying on the presence of any interfaces. We show the complete mapping of the T-spin of surface modes to that of the bulk modes by introducing the coupling between electric and magnetic responses along orthogonal directions, i.e., the bianisotropy, into the medium. We further discover that an interface formed by two bianisotropic media of opposite orientations supports edge-dependent propagating modes with tunable cutoff frequencies. Our results provide a new platform for manipulating the spin orbit interaction of electromagnetic waves.

physics.optics↗

arXiv2018

Giant Asymmetric Radiation from an Ultrathin Bianisotropic Metamaterial

Unidirectional radiation is of particular interest in high-power lasing and optics. Commonly, however, it is difficult to achieve a unidirectional profile in such a system without breaking reciprocity. Recently, assisted by metamaterials without structural symmetry, antennas that radiate asymmetrically have been developed, hence providing the possibility of achieving unidirectionality. Nevertheless, it has been challenging to achieve extremely high radiation asymmetry in such antennas. Here, we demonstrate that this radiation asymmetry is further enhanced when magnetic plasmons are present in the metamaterials. Experimentally, we show that a thin metamaterial with a thickness of approximately λ_0/8 can exhibit a forward-to-backward emission asymmetry of up to 1:32 without any optimization. Our work paves the way for manipulating asymmetric radiation by means of metamaterials and may have a variety of promising applications, such as directional optical and quantum emitters, lasers, and absorbers.

physics.app-ph↗

arXiv2015

Dynamic Bivariate Normal Copula

Normal copula with a correlation coefficient between $-1$ and $1$ is tail independent and so it severely underestimates extreme probabilities. By letting the correlation coefficient in a normal copula depend on the sample size, Hüsler and Reiss (1989) showed that the tail can become asymptotically dependent. In this paper, we extend this result by deriving the limit of the normalized maximum of $n$ independent observations, where the $i$-th observation follows from a normal copula with its correlation coefficient being either a parametric or a nonparametric function of $i/n$. Furthermore, both parametric and nonparametric inference for this unknown function are studied, which can be employed to test the condition in Hüsler and Reiss (1989). A simulation study and real data analysis are presented too.

stat.ME↗

arXiv2014

Test for a Mean Vector with Fixed or Divergent Dimension

It has been a long history in testing whether a mean vector with a fixed dimension has a specified value. Some well-known tests include the Hotelling $T^2$-test and the empirical likelihood ratio test proposed by Owen [Biometrika 75 (1988) 237-249; Ann. Statist. 18 (1990) 90-120]. Recently, Hotelling $T^2$-test has been modified to work for a high-dimensional mean, and the empirical likelihood method for a mean has been shown to be valid when the dimension of the mean vector goes to infinity. However, the asymptotic distributions of these tests depend on whether the dimension of the mean vector is fixed or goes to infinity. In this paper, we propose to split the sample into two parts and then to apply the empirical likelihood method to two equations instead of d equations, where d is the dimension of the underlying random vector. The asymptotic distribution of the new test is independent of the dimension of the mean vector. A simulation study shows that the new test has a very stable size with respect to the dimension of the mean vector, and is much more powerful than the modified Hotelling $T^2$-test.

stat.ME↗

arXiv2014

Inference for a Special Bilinear Time Series Model

It is well known that estimating bilinear models is quite challenging. Many different ideas have been proposed to solve this problem. However, there is not a simple way to do inference even for its simple cases. This paper studies the special bilinear model $$Y_t=μ+ϕY_{t-2}+ bY_{t-2}\varepsilon_{t-1}+ \varepsilon_t,$$ where $\{\varepsilon_t\}$ is a sequence of i.i.d. random variables with mean zero. We first give a sufficient condition for the existence of a unique stationary solution for the model and then propose a GARCH-type maximum likelihood estimator for estimating the unknown parameters. It is shown that the GMLE is consistent and asymptotically normal under only finite fourth moment of errors. Also a simple consistent estimator for the asymptotic covariance is provided. A simulation study confirms the good finite sample performance. Our estimation approach is novel and nonstandard and it may provide a new insight for future research in this direction.

math.ST↗

arXiv2014

Predictive regressions for macroeconomic data

Researchers have constantly asked whether stock returns can be predicted by some macroeconomic data. However, it is known that macroeconomic data may exhibit nonstationarity and/or heavy tails, which complicates existing testing procedures for predictability. In this paper we propose novel empirical likelihood methods based on some weighted score equations to test whether the monthly CRSP value-weighted index can be predicted by the log dividend-price ratio or the log earnings-price ratio. The new methods work well both theoretically and empirically regardless of the predicting variables being stationary or nonstationary or having an infinite variance.

stat.AP↗

arXiv2014

Maxima of a triangular array of multivariate Gaussian sequence

It is known that the normalized maxima of a sequence of independent and identically distributed bivariate normal random vectors with correlation coefficient $ρ\in (-1,1)$ is asymptotically independent, which may seriously underestimate extreme probabilities in practice. By letting $ρ$ depend on the sample size and go to one with certain rate, Hüsler and Reiss (1989) showed that the normalized maxima can become asymptotically dependent. In this paper, we extend such a study to a triangular array of multivariate Gaussian sequence, which further generalizes the results in Hsing, Hüsler and Reiss (1996) and Hashorva and Weng (2013).

math.PR↗

arXiv2013

Estimation of Extreme Quantiles for Functions of Dependent Random Variables

We propose a new method for estimating the extreme quantiles for a function of several dependent random variables. In contrast to the conventional approach based on extreme value theory, we do not impose the condition that the tail of the underlying distribution admits an approximate parametric form, and, furthermore, our estimation makes use of the full observed data. The proposed method is semiparametric as no parametric forms are assumed on all the marginal distributions. But we select appropriate bivariate copulas to model the joint dependence structure by taking the advantage of the recent development in constructing large dimensional vine copulas. Consequently a sample quantile resulted from a large bootstrap sample drawn from the fitted joint distribution is taken as the estimator for the extreme quantile. This estimator is proved to be consistent. The reliable and robust performance of the proposed method is further illustrated by simulation.

stat.ME↗

arXiv2013

Tests for covariance matrix with fixed or divergent dimension

Testing covariance structure is of importance in many areas of statistical analysis, such as microarray analysis and signal processing. Conventional tests for finite-dimensional covariance cannot be applied to high-dimensional data in general, and tests for high-dimensional covariance in the literature usually depend on some special structure of the matrix. In this paper, we propose some empirical likelihood ratio tests for testing whether a covariance matrix equals a given one or has a banded structure. The asymptotic distributions of the new tests are independent of the dimension.

math.ST↗

arXiv2013

Weighted estimation of the dependence function for an extreme-value distribution

Bivariate extreme-value distributions have been used in modeling extremes in environmental sciences and risk management. An important issue is estimating the dependence function, such as the Pickands dependence function. Some estimators for the Pickands dependence function have been studied by assuming that the marginals are known. Recently, Genest and Segers [Ann. Statist. 37 (2009) 2990-3022] derived the asymptotic distributions of those proposed estimators with marginal distributions replaced by the empirical distributions. In this article, we propose a class of weighted estimators including those of Genest and Segers (2009) as special cases. We propose a jackknife empirical likelihood method for constructing confidence intervals for the Pickands dependence function, which avoids estimating the complicated asymptotic variance. A simulation study demonstrates the effectiveness of our proposed jackknife empirical likelihood method.

math.ST↗

arXiv2013

Parameter estimation and model testing for Markov processes via conditional characteristic functions

Markov processes are used in a wide range of disciplines, including finance. The transition densities of these processes are often unknown. However, the conditional characteristic functions are more likely to be available, especially for Lévy-driven processes. We propose an empirical likelihood approach, for both parameter estimation and model specification testing, based on the conditional characteristic function for processes with either continuous or discontinuous sample paths. Theoretical properties of the empirical likelihood estimator for parameters and a smoothed empirical likelihood ratio test for a parametric specification of the process are provided. Simulations and empirical case studies are carried out to confirm the effectiveness of the proposed estimator and test.

math.ST↗

arXiv2011

The scattering of a cylindrical invisibility cloak: reduced parameters and optimization

We investigate the scattering of 2D cylindrical invisibility cloaks with simplified constitutive parameters with the assistance of scattering coefficients. We show that the scattering of the cloaks originates not only from the boundary conditions but also from the spatial variation of the component of permittivity/permeability. According to our formulation, we propose some restrictions to the invisibility cloak in order to minimize its scattering after the simplification has taken place. With our theoretical analysis, it is possible to design a simplified cloak by using some peculiar composites like photonic crystals (PCs) which mimic an effective refractive index landscape rather than offering effective constitutives, meanwhile canceling the scattering from the inner and outer boundaries.

physics.optics↗

arXiv2010

Nearly-zero transmission through periodically modulated ultrathin metal films

Transmission of light through an optically ultrathin metal film with a thickness comparable to its skin depth is significant. We demonstrate experimentally nearly-zero transmission of light through a film periodically modulated by a one-dimensional array of subwavelength slits. The suppressed optical transmission is due to the excitation of surface plasmon polaritons and the zero-transmission phenomenon is strongly dependent on the polarization of the incident wave.

physics.optics↗

arXiv2010

Achieving Anisotropy in Metamaterials made of Dielectric Cylindrical Rods

We show that anisotropic negative effective dispersion relation can be achieved in pure dielectric rod-type metamaterials by turning from the symmetry of a square lattice to that of a rectangular one, i.e. by breaking the rotation symmetry of effective homogeneous medium. Theoretical predictions and conclusions are verified by both numerical calculations and computer based simulations. The proposed anisotropic metamaterial, is used to construct a refocusing slab-lens and a subdiffraction hyperlens. The all-dielectric origin makes it more straightforward to address loss and scaling, two major issues of metallic structures, thus facilitating future applications in both the terahertz and optical range.

physics.optics↗

arXiv2009

Enhanced transmission of transverse electric waves through periodic arrays of structured subwavelength apertures

Transmission through sub-wavelength apertures in perfect metals is expected to be strongly suppressed. However, by structural engineering of the apertures, we numerically demonstrate that the transmission of transverse electric waves through periodic arrays of subwavelength apertures in a thin metallic film can be significantly enhanced. Based on equivalent circuit theory analysis, periodic arrays of square structured subwavelength apertures are obtained with a 1900-fold transmission enhancement factor when the side length $a$ of the apertures is 10 times smaller than the wavelength ($a/λ=0.1$). By examining the induced surface currents and investigating the influence of the lattice constant and the incident angle to the resonant frequency, we show that the enhancement is due to the excitation of the strong localized resonant modes of the structured apertures.

physics.optics↗

arXiv2009

Sub-diffraction-limit Observation Realized by Nonlinear Metamaterial Lens

In this paper, we show by experiment that by covering a thin flat nonlinear lens on the sources, the sub-diffraction-limit observation can be achieved by measuring either the near-field distribution or the far-field radiation of the sources at the harmonic frequencies and calculating the inverse Fourier transformation to obtain the sub-wavelength imaging. Especially, the sub-wavelength image calculated from measured far-field data demonstrates very clear resolution. Since metamaterials included with active elements can easily behave strong nonlinearity under very weak incident electromagnetic powers, the application of the nonlinear lens proposed in this paper would have important potential in improving the sub-wavelength resolution in the near future.

physics.optics↗