arXiv subjects

Chen Liu

Publications and source records attributed to Chen Liu.

At least 271 records · Page 15Linked to original sources

arXiv2018

An interior penalty discontinuous Galerkin approach for 3D incompressible Navier--Stokes equation for permeability estimation of porous media

Permeability estimation of porous media from direct solving Navier--Stokes equation has a wide spectrum of applications in petroleum industry. In this paper, we utilize a pressure-correction projection algorithm in conjunction with the interior penalty discontinuous Galerkin scheme for space discretization to build an incompressible Navier--Stokes simulator and to use this simulator to calculate permeability of real rock sample. The proposed method is accurate, numerically robust, and exhibits the potential for tackling realistic problems.

physics.comp-ph↗

arXiv2018

PlaneNet: Piece-wise Planar Reconstruction from a Single RGB Image

This paper proposes a deep neural network (DNN) for piece-wise planar depthmap reconstruction from a single RGB image. While DNNs have brought remarkable progress to single-image depth prediction, piece-wise planar depthmap reconstruction requires a structured geometry representation, and has been a difficult task to master even for DNNs. The proposed end-to-end DNN learns to directly infer a set of plane parameters and corresponding plane segmentation masks from a single RGB image. We have generated more than 50,000 piece-wise planar depthmaps for training and testing from ScanNet, a large-scale RGBD video database. Our qualitative and quantitative evaluations demonstrate that the proposed approach outperforms baseline methods in terms of both plane segmentation and depth estimation accuracy. To the best of our knowledge, this paper presents the first end-to-end neural architecture for piece-wise planar reconstruction from a single RGB image. Code and data are available at https://github.com/art-programmer/PlaneNet.

cs.CV↗

arXiv2018

FloorNet: A Unified Framework for Floorplan Reconstruction from 3D Scans

The ultimate goal of this indoor mapping research is to automatically reconstruct a floorplan simply by walking through a house with a smartphone in a pocket. This paper tackles this problem by proposing FloorNet, a novel deep neural architecture. The challenge lies in the processing of RGBD streams spanning a large 3D space. FloorNet effectively processes the data through three neural network branches: 1) PointNet with 3D points, exploiting the 3D information; 2) CNN with a 2D point density image in a top-down view, enhancing the local spatial reasoning; and 3) CNN with RGB images, utilizing the full image information. FloorNet exchanges intermediate features across the branches to exploit the best of all the architectures. We have created a benchmark for floorplan reconstruction by acquiring RGBD video streams for 155 residential houses or apartments with Google Tango phones and annotating complete floorplan information. Our qualitative and quantitative evaluations demonstrate that the fusion of three branches effectively improves the reconstruction quality. We hope that the paper together with the benchmark will be an important step towards solving a challenging vector-graphics reconstruction problem. Code and data are available at https://github.com/art-programmer/FloorNet.

cs.CV↗

arXiv2018

Teaching Autonomous Driving Using a Modular and Integrated Approach

Autonomous driving is not one single technology but rather a complex system integrating many technologies, which means that teaching autonomous driving is a challenging task. Indeed, most existing autonomous driving classes focus on one of the technologies involved. This not only fails to provide a comprehensive coverage, but also sets a high entry barrier for students with different technology backgrounds. In this paper, we present a modular, integrated approach to teaching autonomous driving. Specifically, we organize the technologies used in autonomous driving into modules. This is described in the textbook we have developed as well as a series of multimedia online lectures designed to provide technical overview for each module. Then, once the students have understood these modules, the experimental platforms for integration we have developed allow the students to fully understand how the modules interact with each other. To verify this teaching approach, we present three case studies: an introductory class on autonomous driving for students with only a basic technology background; a new session in an existing embedded systems class to demonstrate how embedded system technologies can be applied to autonomous driving; and an industry professional training session to quickly bring up experienced engineers to work in autonomous driving. The results show that students can maintain a high interest level and make great progress by starting with familiar concepts before moving onto other modules.

cs.CY↗

arXiv2018

Mean-field scenario for the athermal creep dynamics of yield-stress fluids

We develop an elasto-plastic description for the transient dynamics prior to steady flow of athermally yielding materials. Our mean-field model not only reproduces the experimentally observed non-linear time dependence of the shear-rate response to an external shear-stress, but also allows for the determination of the different physical processes involved in the onset of the re-acceleration phase after the initial critical slowing down and a distinct well defined fluidization phase. The evidenced power-law dependence of the fluidization time on the distance of the applied to an age dependent static yield stress is not universal but strongly dependent on initial conditions.

cond-mat.soft↗

arXiv2017

An energy-based equilibrium contact angle boundary condition on jagged surfaces for phase-field methods

We consider an energy-based boundary condition to impose an equilibrium wetting angle for the Cahn-Hilliard-Navier-Stokes phase-field model on voxel-set-type computational domains. These domains typically stem from the micro-CT imaging of porous rock and approximate a (on μm scale) smooth domain with a certain resolution. Planar surfaces that are perpendicular to the main axes are naturally approximated by a layer of voxels. However, planar surfaces in any other directions and curved surfaces yield a jagged/rough surface approximation by voxels. For the standard Cahn-Hilliard formulation, where the contact angle between the diffuse interface and the domain boundary (fluid-solid interface/wall) is 90 degrees, jagged surfaces have no impact on the contact angle. However, a prescribed contact angle smaller or larger than 90 degrees on jagged voxel surfaces is amplified in either direction. As a remedy, we propose the introduction of surface energy correction factors for each fluid-solid voxel face that counterbalance the difference of the voxel-set surface area with the underlying smooth one. The discretization of the model equations is performed with the discontinuous Galerkin method, however, the presented semi-analytical approach of correcting the surface energy is equally applicable to other direct numerical methods such as finite elements, finite volumes, or finite differences, since the correction factors appear in the strong formulation of the model.

physics.flu-dyn↗

arXiv2017

Label-free three-dimensional (3D) structural imaging in live cells using intrinsic optical refractive index

Here we report a method for visualization of volumetric structural information of live biological samples with no exogenous contrast agents. The process is made possible through a technique that involves generation, synthesis and analysis of three-dimensional (3D) Fourier components of light diffracted by the sample. This leads to the direct recovery of quantitative cellular morphology with no iterative procedures for reduced computational complexity. Combing with the fact that the technique is easily adaptive to any imaging platform and requires minimum sample preparation, our proposed method is particularly promising for observing fast, volumetric and dynamic events previously only accessible through staining methods.

q-bio.QM↗

arXiv2016

Capacity and Delay Tradeoff of Secondary Cellular Networks with Spectrum Aggregation

Cellular communication networks are plagued with redundant capacity, which results in low utilization and cost-effectiveness of network capital investments. The redundant capacity can be exploited to deliver secondary traffic that is ultra-elastic and delay-tolerant. In this paper, we propose an analytical framework to study the capacity-delay tradeoff of elastic/secondary traffic in large scale cellular networks with spectrum aggregation. Our framework integrates stochastic geometry and queueing theory models and gives analytical insights into the capacity-delay performance in the interference limited regime. Closed-form results are obtained to characterize the mean delay and delay distribution as functions of per user throughput capacity. The impacts of spectrum aggregation, user and base station (BS) densities, traffic session payload, and primary traffic dynamics on the capacity-delay tradeoff relationship are investigated. The fundamental capacity limit is derived and its scaling behavior is revealed. Our analysis shows the feasibility of providing secondary communication services over cellular networks and highlights some critical design issues.

cs.NI↗

arXiv2016

Deep Multi-Modal Image Correspondence Learning

Inference of correspondences between images from different modalities is an extremely important perceptual ability that enables humans to understand and recognize cross-modal concepts. In this paper, we consider an instance of this problem that involves matching photographs of building interiors with their corresponding floorplan. This is a particularly challenging problem because a floorplan, as a stylized architectural drawing, is very different in appearance from a color photograph. Furthermore, individual photographs by themselves depict only a part of a floorplan (e.g., kitchen, bathroom, and living room). We propose the use of a number of different neural network architectures for this task, which are trained and evaluated on a novel large-scale dataset of 5 million floorplan images and 80 million associated photographs. Experimental evaluation reveals that our neural network architectures are able to identify visual cues that result in reliable matches across these two quite different modalities. In fact, the trained networks are able to even outperform human subjects in several challenging image matching problems. Our result implies that neural networks are effective at perceptual tasks that require long periods of reasoning even for humans to solve.

cs.CV↗

arXiv2016

Multi-way Particle Swarm Fusion

This paper proposes a novel MAP inference framework for Markov Random Field (MRF) in parallel computing environments. The inference framework, dubbed Swarm Fusion, is a natural generalization of the Fusion Move method. Every thread (in a case of multi-threading environments) maintains and updates a solution. At each iteration, a thread can generate arbitrary number of solution proposals and take arbitrary number of concurrent solutions from the other threads to perform multi-way fusion in updating its solution. The framework is general, making popular existing inference techniques such as alpha-expansion, fusion move, parallel alpha-expansion, and hierarchical fusion, its special cases. We have evaluated the effectiveness of our approach against competing methods on three problems of varying difficulties, in particular, the stereo, the optical flow, and the layered depthmap estimation problems.

cs.CV↗

arXiv2016

A finite volume/discontinuous Galerkin method for the advective Cahn-Hilliard equation with degenerate mobility on porous domains stemming from micro-CT imaging

A numerical method is formulated for the solution of the advective Cahn-Hilliard (CH) equation with constant and degenerate mobility in three-dimensional porous media with non-vanishing velocity on the exterior boundary. The CH equation describes phase separation of an immiscible binary mixture at constant temperature in the presence of a mass constraint and dissipation of free energy. Porous media/pore-scale problems specifically entail high-resolution images of rocks in which the solid matrix and pore spaces are fully resolved. The interior penalty discontinuous Galerkin method is used for the spatial discretization of the CH equation in mixed form, while a semi-implicit convex-concave splitting is utilized for temporal discretization. The spatial approximation order is arbitrary, while it reduces to a finite volume scheme for the choice of elementwise constants. The resulting nonlinear systems of equations are reduced using the Schur complement and solved via Newton's method. The numerical scheme is first validated using numerical convergence tests and then applied to a number of fundamental problems for validation and numerical experimentation purposes including the case of degenerate mobility. First-order physical applicability and robustness of the numerical method are shown in a breakthrough scenario on a voxel set obtained from a micro-CT scan of a real sandstone rock sample.

math.NA↗

arXiv2016

Electromagnetic transitions in multiple chiral doublet bands

Multiple chiral doublet bands (M$χ$D) in the $80$, 130 and $190$ mass regions are studied by the model of $γ$=90$^{\circ}$ triaxial rotor coupled with identical symmetric proton-neutron configurations. By selecting the suitable basis, the calculated wave functions are explicitly exhibited to be symmetric under the operator $\hat{A}$, which is defined as rotation by $90^{\circ}$ about 3-axis with the exchange of valance proton and neutron. We found that both $M1$ and $E2$ transitions are allowed between the levels with different values of $A$, while are forbidden between the levels with same values of $A$. Such a selection rule holds true for M$χ$D in different mass regions.

nucl-th↗

arXiv2016

Modeling of a Single Multimode Fiber Imaging System

We present a detailed theoretical analysis on image transmission via a single multimode fiber (MMF). A single MMF imaging model is developed to study the light wave propagation from the light source to the camera, by using free-space Fourier optics theory and mode-coupling theory. A mathematical expression is obtained for the complete single MMF imaging system, which is further validated by image-transmission simulations. Our model is believed to be the first theoretical model to describe the complete MMF imaging system based on the transmission of individual modes. Therefore, this model is robust and capable of analyzing MMF image transmission under specific mode-coupling conditions. We use our model to study bending-induced image blur in single-MMF image transmission, and the result has found a good agreement with that of existing experimental studies. These should provide important insights into future MMF imaging system developments.

physics.optics↗

arXiv2016

Generation of Gamma-ray Beam with Orbital Angular Momentum in the QED Regime

We propose a scheme to generate gamma-ray photons with an orbital angular momentum (OAM) and high energy simultaneously from laser-plasma interactions by irradiating a circularly polarized Laguerre-Gaussian laser on a thin plasma target. The spin angular momentum and OAM are first transferred to electrons from the driving laser photons, and then the OAM is transferred to the gamma-ray photons from the electrons through quantum radiation. This scheme has been demonstrated using three-dimensional quantum electrodynamics particle-in-cell simulation. The topological charge, chirality and carrier-envelope phase of the short ultra-intense vortex laser can be revealed according to the pattern feature of the energy density of radiated photons.

physics.plasm-ph↗

arXiv2015

Driving rate dependence of avalanche statistics and shapes at the yielding transition

We study stress time series caused by plastic avalanches in athermally sheared disordered materials. Using particle-based simulations and a mesoscopic elasto-plastic model, we analyze size and shear-rate dependence of the stress-drop durations and size distributions together with their average temporal shape. We find critical exponents different from mean-field predictions, and a clear asymmetry for individual avalanches. We probe scaling relations for the rate dependency of the dynamics and we report a crossover towards mean-field results for strong driving.

cond-mat.soft↗

arXiv2015

Nature of the Insulating Ground State of the 5d PostPerovskite CaIrO3

The insulating ground state of the 5d transition metal oxide CaIrO3 has been classified as a Mott-type insulator. Based on a systematic density functional theory (DFT) study with local, semilocal, and hybrid exchange-correlation functionals, we reveal that the Ir t2g states exhibit large splittings and one-dimensional electronic states along the c axis due to a tetragonal crystal field. Our hybrid DFT calculation adequately describes the antiferromagnetic (AFM) order along the c direction via a superexchange interaction between Ir4+ spins. Furthermore, the spin-orbit coupling (SOC) hybridizes the t2g states to open an insulating gap. These results indicate that CaIrO3 can be represented as a spin-orbit Slater insulator, driven by the interplay between a long-range AFM order and the SOC. Such a Slater mechanism for the gap formation is also demonstrated by the DFT + dynamical mean field theory calculation, where the metal-insulator transition and the paramagnetic to AFM phase transition are concomitant with each other.

cond-mat.str-el↗

arXiv2015

Generation of Ultra-intense Gamma-ray Train by QED Harmonics

When laser intensity exceeds 10^22W/cm^2, photons with energy above MeV can be generated from high-order harmonics process in the laser-plasma interaction. We find that under such laser intensity, QED effect plays a dominating role in the radiation pattern. Contrast to the gas and relativistic HHG processes, both the occurrence and energy of gamma-ray emission produced by QED harmonics are random and QED harmonics are usually not coherent, while the property of high intensity and ultra-short duration is conserved. Our simulation shows that the period of gamma-ray train is half of the laser period and the peak intensity is 1.4e22W/cm^2. This new harmonic production with QED effects are crucial to light-matter interaction in strong field and can be verified in experiments by 10PW laser facilities in the near future.

physics.plasm-ph↗

arXiv2015

The Open Access Advantage Considering Citation, Article Usage and Social Media Attention

In this study, we compare the difference in the impact between open access (OA) and non-open access (non-OA) articles. 1761 Nature Communications articles published from 1 Jan. 2012 to 31 Aug. 2013 are selected as our research objects, including 587 OA articles and 1174 non-OA articles. Citation data and daily updated article-level metrics data are harvested directly from the platform of nature.com. Data is analyzed from the static versus temporal-dynamic perspectives. The OA citation advantage is confirmed, and the OA advantage is also applicable when extending the comparing from citation to article views and social media attention. More important, we find that OA papers not only have the great advantage of total downloads, but also have the feature of keeping sustained and steady downloads for a long time. For article downloads, non-OA papers only have a short period of attention, when the advantage of OA papers exists for a much longer time.

cs.DL↗