Searcharxiv⌕ Search

arXiv subjects

Qi Chen

Publications and source records attributed to Qi Chen.

At least 307 records · Page 17Linked to original sources
arXiv2019

Cooper: Cooperative Perception for Connected Autonomous Vehicles based on 3D Point Clouds

Autonomous vehicles may make wrong decisions due to inaccurate detection and recognition. Therefore, an intelligent vehicle can combine its own data with that of other vehicles to enhance perceptive ability, and thus improve detection accuracy and driving safety. However, multi-vehicle cooperative perception requires the integration of real world scenes and the traffic of raw sensor data exchange far exceeds the bandwidth of existing vehicular networks. To the best our knowledge, we are the first to conduct a study on raw-data level cooperative perception for enhancing the detection ability of self-driving systems. In this work, relying on LiDAR 3D point clouds, we fuse the sensor data collected from different positions and angles of connected vehicles. A point cloud based 3D object detection method is proposed to work on a diversity of aligned point clouds. Experimental results on KITTI and our collected dataset show that the proposed system outperforms perception by extending sensing area, improving detection accuracy and promoting augmented results. Most importantly, we demonstrate it is possible to transmit point clouds data for cooperative perception via existing vehicular network technologies.

cs.CV↗
arXiv2019

Auto-Embedding Generative Adversarial Networks for High Resolution Image Synthesis

Generating images via the generative adversarial network (GAN) has attracted much attention recently. However, most of the existing GAN-based methods can only produce low-resolution images of limited quality. Directly generating high-resolution images using GANs is nontrivial, and often produces problematic images with incomplete objects. To address this issue, we develop a novel GAN called Auto-Embedding Generative Adversarial Network (AEGAN), which simultaneously encodes the global structure features and captures the fine-grained details. In our network, we use an autoencoder to learn the intrinsic high-level structure of real images and design a novel denoiser network to provide photo-realistic details for the generated images. In the experiments, we are able to produce 512x512 images of promising quality directly from the input noise. The resultant images exhibit better perceptual photo-realism, i.e., with sharper structure and richer details, than other baselines on several datasets, including Oxford-102 Flowers, Caltech-UCSD Birds (CUB), High-Quality Large-scale CelebFaces Attributes (CelebA-HQ), Large-scale Scene Understanding (LSUN) and ImageNet.

cs.CV↗
arXiv2018

Ab initio simulation of $\mathrm{Ta_2O_5}$: A high symmetry ground state phase with application to interface calculation

We suggest a tetragonal $I4_1/amd$ phase ($η$-phase) as the ground state of $\mathrm{Ta_2O_5}$ at zero temperature, which is a high symmetry version of the triclinic $γ$-phase $\mathrm{Ta_2O_5}$ predicted by Yang and Kawazoe. Our calculation shows that $γ$-phase $\mathrm{Ta_2O_5}$ will automatically be transformed into the $η$-phase during structural relaxation. Phonon dispersion confirms that the $η$-phase is dynamically stable, while the high temperature $α$-phase $\mathrm{Ta_2O_5}$, which also has the $I4_1/amd$ symmetry, is unstable at zero temperature. A thorough energy comparison of the $β_{AL}$, $δ$, $λ$, $\mathrm{B}$, $\mathrm{L_{SR}}$, $β_R$, $Pm$, $Cmmm$, $γ$, $η$ and $α$ phases of $\mathrm{Ta_2O_5}$ is carried out. The GGA-1/2 method is applied in calculating the electronic structure of various phases, where the $η$-phase demonstrates a 4.24 eV indirect band gap, close to experimental value. The high symmetry tetragonal phase together with computationally efficient GGA-1/2 method greatly facilitate the $ab\ initio$ simulation of $\mathrm{Ta_2O_5}$-based devices. As an example, we have explicitly shown the Ohmic contact nature between metal Ta and $\mathrm{Ta_2O_5}$ by calculating an interface model of $b.c.c.$ Ta and $η$-$\mathrm{Ta_2O_5}$, using GGA-1/2.

cond-mat.mtrl-sci↗
arXiv2018

A 16-pixel NbN nanowire single photon detector coupled with 300 micrometer fiber

Niobium Nitride (NbN) nanowire is the most popular detection material of superconducting nanowire single photon detectors (SNSPDs) for high repetition rate and high efficiency. However, it has been assumed to be difficult for fabricating SNSPDs with arrays over large area, which are critical components of quantum imaging, linear optical quantum computing, satellite laser ranging, and high-speed quantum key distribution. This paper reported a 4*4 pixel NbN SNSPDs array with an equivalent receiving aperture of 300 micrometer associated with beam compression technology, which is the first NbN SNSPD coupled by 300 micrometer fiber according to our knowledge. The designed pixel was compact with a filling factor of 98.5%, resulting in a high quantum efficiency of 94.5%, a system efficiency of 46% for photons coupled from 300 micrometer fiber without optimizing polarization, and a system time resolution of 92 ps. This work demonstrates the feasibility of a large SNSPDs array, and paves the way for developing efficient photon camera with NbN nanowires.

physics.app-ph↗
arXiv2018

DUNet: A deformable network for retinal vessel segmentation

Automatic segmentation of retinal vessels in fundus images plays an important role in the diagnosis of some diseases such as diabetes and hypertension. In this paper, we propose Deformable U-Net (DUNet), which exploits the retinal vessels' local features with a U-shape architecture, in an end to end manner for retinal vessel segmentation. Inspired by the recently introduced deformable convolutional networks, we integrate the deformable convolution into the proposed network. The DUNet, with upsampling operators to increase the output resolution, is designed to extract context information and enable precise localization by combining low-level feature maps with high-level ones. Furthermore, DUNet captures the retinal vessels at various shapes and scales by adaptively adjusting the receptive fields according to vessels' scales and shapes. Three public datasets DRIVE, STARE and CHASE_DB1 are used to train and test our model. Detailed comparisons between the proposed network and the deformable neural network, U-Net are provided in our study. Results show that more detailed vessels are extracted by DUNet and it exhibits state-of-the-art performance for retinal vessel segmentation with a global accuracy of 0.9697/0.9722/0.9724 and AUC of 0.9856/0.9868/0.9863 on DRIVE, STARE and CHASE_DB1 respectively. Moreover, to show the generalization ability of the DUNet, we used another two retinal vessel data sets, one is named WIDE and the other is a synthetic data set with diverse styles, named SYNTHE, to qualitatively and quantitatively analyzed and compared with other methods. Results indicates that DUNet outperforms other state-of-the-arts.

cs.CV↗
arXiv2018

Semantic Segmentation for Urban Planning Maps based on U-Net

The automatic digitizing of paper maps is a significant and challenging task for both academia and industry. As an important procedure of map digitizing, the semantic segmentation section mainly relies on manual visual interpretation with low efficiency. In this study, we select urban planning maps as a representative sample and investigate the feasibility of utilizing U-shape fully convolutional based architecture to perform end-to-end map semantic segmentation. The experimental results obtained from the test area in Shibuya district, Tokyo, demonstrate that our proposed method could achieve a very high Jaccard similarity coefficient of 93.63% and an overall accuracy of 99.36%. For implementation on GPGPU and cuDNN, the required processing time for the whole Shibuya district can be less than three minutes. The results indicate the proposed method can serve as a viable tool for urban planning map semantic segmentation task with high accuracy and efficiency.

cs.LG↗
arXiv2018

Dual Reconstruction Nets for Image Super-Resolution with Gradient Sensitive Loss

Deep neural networks have exhibited promising performance in image super-resolution (SR) due to the power in learning the non-linear mapping from low-resolution (LR) images to high-resolution (HR) images. However, most deep learning methods employ feed-forward architectures, and thus the dependencies between LR and HR images are not fully exploited, leading to limited learning performance. Moreover, most deep learning based SR methods apply the pixel-wise reconstruction error as the loss, which, however, may fail to capture high-frequency information and produce perceptually unsatisfying results, whilst the recent perceptual loss relies on some pre-trained deep model and they may not generalize well. In this paper, we introduce a mask to separate the image into low- and high-frequency parts based on image gradient magnitude, and then devise a gradient sensitive loss to well capture the structures in the image without sacrificing the recovery of low-frequency content. Moreover, by investigating the duality in SR, we develop a dual reconstruction network (DRN) to improve the SR performance. We provide theoretical analysis on the generalization performance of our method and demonstrate its effectiveness and superiority with thorough experiments.

cs.CV↗
arXiv2018

UnrealStereo: Controlling Hazardous Factors to Analyze Stereo Vision

A reliable stereo algorithm is critical for many robotics applications. But textureless and specular regions can easily cause failure by making feature matching difficult. Understanding whether an algorithm is robust to these hazardous regions is important. Although many stereo benchmarks have been developed to evaluate performance, it is hard to quantify the effect of hazardous regions in real images because the location and severity of these regions are unknown. In this paper, we develop a synthetic image generation tool enabling to control hazardous factors, such as making objects more specular or transparent, to produce hazardous regions at different degrees. The densely controlled sampling strategy in virtual worlds enables to effectively stress test stereo algorithms by varying the types and degrees of the hazard. We generate a large synthetic image dataset with automatically computed hazardous regions and analyze algorithms on these regions. The observations from synthetic images are further validated by annotating hazardous regions in real-world datasets Middlebury and KITTI (which gives a sparse sampling of the hazards). Our synthetic image generation tool is based on a game engine Unreal Engine 4 and will be open-source along with the virtual scenes in our experiments. Many publicly available realistic game contents can be used by our tool to provide an enormous resource for development and evaluation of algorithms.

cs.CV↗
arXiv2018

Transition threshold for the 2-D Couette flow in a finite channel

In this paper, we study the transition threshold problem for the 2-D Navier-Stokes equations around the Couette flow $(y,0)$ at large Reynolds number $Re$ in a finite channel. We develop a systematic method to establish the resolvent estimates of the linearized operator and space-time estimates of the linearized Navier-Stokes equations. In particular, three kinds of important effects: enhanced dissipation, inviscid damping and boundary layer, are integrated into the space-time estimates in a sharp form. As an application, we prove that if the initial velocity $v_0$ satisfies $\|v_0-(y, 0)\|_{H^2}\le cRe^{-\frac 12}$ for some small $c$ independent of $Re$, then the solution of the 2-D Navier-Stokes equations remains within $O(Re^{-\frac 12})$ of the Couette flow for any time.

math.AP↗
arXiv2018

Floquet Hofstadter Butterfly on the Kagome and Triangular Lattices

In this work we use Floquet theory to theoretically study the influence of monochromatic circularly and linearly polarized light on the Hofstadter butterfly---induced by a uniform perpendicular magnetic field--for both the kagome and triangular lattices. In the absence of the laser light, the butterfly has fractal structure with inversion symmetry about magnetic flux $ϕ=1/4$, and reflection symmetry about $ϕ=1/2$. As the system is exposed to an external laser, we find circularly polarized light deforms the butterfly by breaking the mirror symmetry at flux $ϕ=1/2$. By contrast, linearly polarized light deforms the original butterfly while preserving the mirror symmetry at flux $ϕ=1/2$. We find the inversion symmetry is always preserved for both linear and circular polarized light. For linearly polarized light, the Hofstadter butterfly depends on the polarization direction. Further, we study the effect of the laser on the Chern number of lowest band in the off-resonance regime (laser frequency is larger than the bandwidth). For circularly polarized light, we find that low laser intensity will not change the Chern number, but beyond a critical intensity the Chern number will change. For linearly polarized light, the Chern number depends on the polarization direction. Our work highlights the generic features expected for the periodically driven Hofstadter problem on different lattices.

cond-mat.mes-hall↗
arXiv2018

SampleAhead: Online Classifier-Sampler Communication for Learning from Synthesized Data

State-of-the-art techniques of artificial intelligence, in particular deep learning, are mostly data-driven. However, collecting and manually labeling a large scale dataset is both difficult and expensive. A promising alternative is to introduce synthesized training data, so that the dataset size can be significantly enlarged with little human labor. But, this raises an important problem in active vision: given an {\bf infinite} data space, how to effectively sample a {\bf finite} subset to train a visual classifier? This paper presents an approach for learning from synthesized data effectively. The motivation is straightforward -- increasing the probability of seeing difficult training data. We introduce a module named {\bf SampleAhead} to formulate the learning process into an online communication between a {\em classifier} and a {\em sampler}, and update them iteratively. In each round, we adjust the sampling distribution according to the classification results, and train the classifier using the data sampled from the updated distribution. Experiments are performed by introducing synthesized images rendered from ShapeNet models to assist PASCAL3D+ classification. Our approach enjoys higher classification accuracy, especially in the scenario of a limited number of training samples. This demonstrates its efficiency in exploring the infinite data space.

cs.CV↗
arXiv2018

On Information-Theoretic Characterizations of Markov Random Fields and Subfields

Let $X_i, i \in V$ form a Markov random field (MRF) represented by an undirected graph $G = (V,E)$, and $V'$ be a subset of $V$. We determine the smallest graph that can always represent the subfield $X_i, i \in V'$ as an MRF. Based on this result, we obtain a necessary and sufficient condition for a subfield of a Markov tree to be also a Markov tree. When $G$ is a path so that $X_i, i \in V$ form a Markov chain, it is known that the $I$-Measure is always nonnegative and the information diagram assumes a very special structure Kawabata and Yeung (1992). We prove that Markov chain is essentially the only MRF such that the $I$-Measure is always nonnegative. By applying our characterization of the smallest graph representation of a subfield of an MRF, we develop a recursive approach for constructing information diagrams for MRFs. Our work is built on the set-theoretic characterization of an MRF in Yeung, Lee, and Ye (2002).

cs.DM↗
arXiv2017

Floquet band structure of a semi-Dirac system

In this work we use Floquet-Bloch theory to study the influence of circularly and linearly polarized light on two-dimensional band structures with semi-Dirac band touching points, taking the anisotropic nearest neighbor hopping model on the honeycomb lattice as an example. We find circularly polarized light opens a gap and induces a band inversion to create a finite Chern number in the two-band model. By contrast, linearly polarized light can either open up a gap (polarized in the quadratically dispersing direction) or split the semi-Dirac band touching point into two Dirac points (polarized in the linearly dispersing direction) by an amount that depends on the amplitude of the light. Motivated by recent pump-probe experiments, we investigated the non-equilibrium spectral properties and momentum-dependent spin-texture of our model in the Floquet state following a quench in absence of phonons, and in the presence of phonon dissipation that leads to a steady-state independent of the pump protocol. Finally, we make connections to optical measurements by computing the frequency dependence of the longitudinal and transverse optical conductivity for this two-band model. We analyze the various contributions from inter-band transitions and different Floquet modes. Our results suggest strategies for optically controlling band structures and experimentally measuring topological Floquet systems.

cond-mat.mes-hall↗
arXiv2017

The HTL resumed propagators in the light cone gauge

The expression of the HTL resumed gluon propagator in the light cone gauge is derived. In the real time mechanism, using the Mandelstam Leibbrant prescription of $(n\cdot K)^{-1}$, we calculate the transverse and longitudinal parts of the gluon HTL self-energy and prove the transverse and longitudinal parts do not have divergence. We also calculate the quark self energy in the HTL approximation, and find it gauge independent. We analytically calculate the damping rates of the hard quark and gluon with this HTL resumed gluon propagator.

hep-ph↗
arXiv2016

Partition-Symmetrical Entropy Functions

Let $\cal{N}=\{1,\cdots,n\}$. The entropy function $\bf h$ of a set of $n$ discrete random variables $\{X_i:i\in\cal N\}$ is a $2^n$-dimensional vector whose entries are ${\bf{h}}({\cal{A}})\triangleq H(X_{\cal{A}}),\cal{A}\subset{\cal N} $, the (joint) entropies of the subsets of the set of $n$ random variables with $H(X_\emptyset)=0$ by convention. The set of all entropy functions for $n$ discrete random variables, denoted by $Γ^*_n$, is called the entropy function region for $n$. Characterization of $Γ^*_n$ and its closure $\overline{Γ^*_n}$ are well-known open problems in information theory. They are important not only because they play key roles in information theory problems but also they are related to other subjects in mathematics and physics. In this paper, we consider \emph{partition-symmetrical entropy functions}. Let $p=\{\cal{N}_1,\cdots, \cal{N}_t\}$ be a $t$-partition of $\cal N$. An entropy function $\bf h$ is called $p$-symmetrical if for all ${\cal A},{\cal B} \subset {\cal N}$, $\bf{h}({\cal A}) = \bf{h}({\cal B})$ whenever $|{\cal A} \cap {\cal N}_i| = |{\cal B} \cap {\cal N}_i|$, $i = 1, \cdots,t$. The set of all the $p$-symmetrical entropy functions, denoted by $Ψ^*_p$, is called $p$-symmetrical entropy function region. We prove that $\overline{Ψ^*_p}$, the closure of $Ψ^*_p$, is completely characterized by Shannon-type information inequalities if and only if $p$ is the $1$-partition or a $2$-partition with one of its blocks being a singleton. The characterization of the partition-symmetrical entropy functions can be useful for solving some information theory and related problems where symmetry exists in the structure of the problems. Keywords: entropy, entropy function, information inequality, polymatroid.

cs.IT↗
arXiv2016

A Hierarchical Distributed Processing Framework for Big Image Data

This paper introduces an effective processing framework nominated ICP (Image Cloud Processing) to powerfully cope with the data explosion in image processing field. While most previous researches focus on optimizing the image processing algorithms to gain higher efficiency, our work dedicates to providing a general framework for those image processing algorithms, which can be implemented in parallel so as to achieve a boost in time efficiency without compromising the results performance along with the increasing image scale. The proposed ICP framework consists of two mechanisms, i.e. SICP (Static ICP) and DICP (Dynamic ICP). Specifically, SICP is aimed at processing the big image data pre-stored in the distributed system, while DICP is proposed for dynamic input. To accomplish SICP, two novel data representations named P-Image and Big-Image are designed to cooperate with MapReduce to achieve more optimized configuration and higher efficiency. DICP is implemented through a parallel processing procedure working with the traditional processing mechanism of the distributed system. Representative results of comprehensive experiments on the challenging ImageNet dataset are selected to validate the capacity of our proposed ICP framework over the traditional state-of-the-art methods, both in time efficiency and quality of results.

cs.CV↗
arXiv2016

Integrality and Gauge Dependence of Hennings TQFTs

We provide a general construction of integral TQFTs over a general commutative ring, $\mathbf{k}$, starting from a finite Hopf algebra over $\mathbf{k}$ which is Frobenius and double balanced. These TQFTs specialize to the Hennings invariants of the respective doubles on closed 3-manifolds. We show the construction applies to index 2 extensions of the Borel parts of Lusztig's small quantum groups for all simple Lie types, yielding integral TQFTs over the cyclotoic integers for surfaces with boundary. We further establish and compute isomorphisms of TQFT functors constructed from Hopf algebras that are related by a strict gauge transformation in the sense of Drinfeld. Formulas for the natural isomorphisms are given in terms of the gauge twist element. These results are combined and applied to show that the Hennings invariant associated to quantum-$sl_2$ takes values in the cyclotomic integers. Using prior results of Chen et al we infer integrality also of the Witten-Reshetikhin-Turaev $SO(3)$ invariant for rational homology spheres. As opposed to most other approaches the methods described in this article do not invoke calculations of skeins, knots polynomials, or representation theory, but follow a combinatorial construction that uses only the elements and operations of the underlying Hopf algebras.

math.GT↗